science communication - sage publications · science communication an interdisciplinary social...

EDITORCAROL L. ROGERS, College of Journalism

University of Maryland

SENIOR CONTRIBUTING EDITORSLOIS-ELLIN DATTA, Datta AnalysisCHRISTOPHER FOREMAN, JR., University of MarylandS. HOLLY STOCKING, Indiana University

EDITORIAL ASSISTANTKAYLEN D. TUCKER

EDITORIAL ADVISORY BOARD

RICHARD P. BARKE, Georgia Institute of TechnologyROBERT D. BULLARD, Clark Atlanta UniversityPETER J. CAWS, The George Washington UniversitySHARON DUNWOODY, University of Wisconsin–MadisonEDNA EINSIEDEL, University of CalgarySHARON M. FRIEDMAN, Lehigh UniversityJACQUES GAILLARD, International Foundation for ScienceSTEPHEN HILGARTNER, Cornell UniversitySHEILA JASANOFF, Harvard UniversityDAVID H. JOHNSON, Federation of Behavioral, Psychological, and Cognitive SciencesRON JOHNSTON, Australian Centre for Innovation and International CompetitivenessMARCEL C. LAFOLLETTE, Independent ScholarSTEPHEN LOCK, The Wellcome Institute for the History of MedicineROY M. MACLEOD, University of SydneyKATHERINE W. MCCAIN, Drexel UniversityALAN MCGOWAN, Gene Media ForumWILLIAM J. PAISLEY, Knowledge Access InternationalJOHN V. PAVLIK, Columbia UniversityROBERT PERLOFF, University of PittsburghANDREW PICKERING, University of IllinoisSUSANNA HORNIG PRIEST, Texas A&M UniversityDAVID RHEES, The Bakken Library and MuseumROBERT F. RICH, University of IllinoisEVERETT M. ROGERS, University of New MexicoKATHERINE E. ROWAN, George Mason UniversityJURGEN SCHMANDT, University of TexasALBERT H. TEICH, American Association for the Advancement of ScienceTHOMAS S. ULEN, University of IllinoisJOANN MYER VALENTI, Brigham Young UniversityVIVIAN WEIL, Illinois Institute of TechnologyCAROL H. WEISS, Harvard UniversityLEE WILKINS, University of MissouriMARY WOOLLEY, Research! AmericaBRIAN WYNNE, University of Lancaster

For Sage Publications: Stephanie Lawrence, Barbara Corrigan, Muriel Murphy,and Katinka Baltazar

SCIENCECOMMUNICATION

Science CommunicationAn Interdisciplinary Social Science Journal

Volume 23, Number 2December 2001

Special Issue: Understanding Public Communicationof Science and Technology

Contents

EditorialUnderstanding Public Communication

of Science and TechnologyCarol L. Rogers 95

ArticlesMisplaced Faith: Communication Variables as Predictors

of Encouragement for Biotechnology DevelopmentSusanna Hornig Priest 97

Interactivity, Information Processing, and Learningon the World Wide Web

Mark Tremayne and Sharon Dunwoody 111

Science Mass Communication: Its Conceptual HistoryRobert A. Logan 135

Communicating Science: A Review of the LiteratureMichael F. Weigold 164

Communicating the Future: Report of the Research RoadmapPanel for Public Communication of Science andTechnology in the Twenty-first Century

Rick E. Borchelt 194

News and Notices 212

Sage Publications Thousand Oaks • London • New Delhi

SCIENCECOMMUNICATION is an international, interdisciplinary social science journal that exam-ines the nature of expertise, the diffusion of knowledge, and the communication of science andtechnology among professionals and to the public. SCIENCE COMMUNICATION welcomes sub-missions from authors from all disciplines (e.g., social sciences, policy sciences, economics, andhistory) and organizations (e.g., universities, government, and the private sector).Preference is givento articles that bridge the gap between theory and practice and that have application beyond onediscipline. SCIENCE COMMUNICATION also publishes scholarly commentaries, reports, andreviews of all types, including museum exhibits, television, and print media.

MANUSCRIPTSUBMISSIONS:All manuscripts should be submitted in triplicate to Carol L.Rogers,Editor, Science Communication, College of Journalism, University of Maryland, College Park, MD20742-7111. The letter of transmittal should contain (a) the names, addresses, and telephonenumbers of all authors; (b) a statement that all authors have read and approved the manuscriptsubmitted; and (c) a statement that the material has not been previously published and is not underconsideration for publication elsewhere. Each manuscript, typed double-spaced, should include anabstract of not more than 100 words;all tables, figures, footnotes, and alphabetical list of references,also double-spaced, should be appended separately. Citations should be prepared in accord withDocumentation 2 of the Chicago Manual of Style (14th edition, 1993). To facilitate anonymous re-view, names and affiliations of all authors should appear only on the title sheet. Inquiries about sub-missions should be directed to the editor at the address above or by phone: (301) 405-2430; fax: (301)314-9166; or e-mail: [email protected].

SCIENCE COMMUNICATION (ISSN 1075-5470) is published four times annually—in September, December, March,and June—by Sage Publications, 2455 Teller Road, Thousand Oaks, CA 91320; telephone (800) 818-SAGE (7243)and (805) 499-9774; fax/order line (805) 499-0871;e-mail [email protected];http://www.sagepub.com.Copyright© 2001 by Sage Publications. All rights reserved. No portion of the contents may be reproduced in any form with-out written permission of the publisher.

Subscriptions: Annual subscription rates for institutions and individuals are based on the current frequency.Prices quoted are in U.S.dollars and are subject to change without notice.Canadian subscribers add 7% GST (andHST as appropriate). Outside U.S. subscription rates include shipping via air-speeded delivery. Institutions: $455(within the U.S.) / $471 (outside the U.S.) / single issue: $130 (worldwide). Individuals: $94 (within the U.S.) / $110(outside the U.S.) / single issue: $36 (worldwide). Orders from the U.K., Europe, the Middle East, and Africa shouldbe sent to the London address (below). Orders from India and South Asia should be sent to the New Delhi address(below). Noninstitutional orders must be paid by personal check, VISA, or MasterCard.

Periodicals postage paid at Thousand Oaks, California, and at additional mailing addresses.

This journal is abstracted in Abstract Journal of the Educational Resources Information Center,Communication Abstracts,Corporate ResourceNET,Current Citations Express,Current Contents: Social &Behavioral Sciences, Current Index to Journals in Education (CIJE), Environment Abstracts, MasterFILEFullTEXT, Periodical Abstracts, Social Sciences Citation Index, Social Services Abstracts, SociologicalAbstracts, Standard Periodical Directory (SPD), and TOPICsearch and is available on microfilm from UniversityMicrofilms, Ann Arbor, Michigan.

Back Issues: Information about availability and prices of back issues may be obtained from the publisher’s orderdepartment (address below).Single-issue orders for five or more copies will receive a special adoption discount.Con-tact the order department for details. Write to the London office for sterling prices.

All subscription inquiries, orders, and renewals with ship-to addresses in North America, South America, and Canadamust be addressed to Sage Publications, 2455 Teller Road, Thousand Oaks, California 91320, U.S.A., telephone(800) 818-SAGE (7243) and (805) 499-0721, fax (805) 499-0871; e-mail [email protected]; http://www.sagepub.com.All subscription inquiries, orders, and renewals with ship-to addresses in the U.K., Europe, the Middle East, andAfrica must be addressed to Sage Publications Ltd., 6 Bonhill Street, London EC2A 4PU, England, telephone +44(0)20 7374 0645, fax +44 (0)20 7374 8741.All subscriptions inquiries, orders, and renewals with ship-to addresses inIndia and South Asia must be addressed to Sage Publications Private Ltd., P.O.Box 4215, New Delhi 110 048 India,telephone (91-11) 641-9884, fax (91-11) 647-2426.Address all permissions requests to the Thousand Oaks office.

Authorization to photocopy items for internal or personal use, or the internal or personal use of specific clients, isgranted by Sage Publications for libraries and other users registered with the Copyright Clearance Center (CCC)Transactional Reporting Service, provided that the base fee of 50¢ per copy, plus 10¢ per copy page, is paid directlyto CCC, 21 Congress St., Salem, MA 01970. 1075-5470/2001 $.50 + .10.

Advertising: Current rates and specifications may be obtained by writing to the Advertising Manager at theThousand Oaks office (address above).

Claims: Claims for undelivered copies must be made no later than six months following month of publication. Thepublisher will supply missing copies when losses have been sustained in transit and when the reserve stock will permit.

Change of Address:Six weeks’advance notice must be given when notifying of change of address.Please send oldaddress label along with the new address to ensure proper identification.Please specify name of journal.POSTMAS-TER: Send address changes to: Science Communication, c/o 2455 Teller Road, Thousand Oaks, CA 91320.

Printed on acid-free paper

SCIENCE COMMUNICATIONRogers / EDITORIAL

Editorial: UnderstandingPublic Communication

of Science and Technology

A rich literature in the broadly defined areas of science communication, pub-lic communication of science and technology, and public understanding ofscience has developed over the past twenty-five to thirty years, in particular.This scholarly work can be found in specialized journals such as ScienceCommunication; research articles in other journals devoted to both scienceand communication; sessions at conferences that focus on science and oncommunication research; specialized conferences such as the InternationalConference on Science Communication in Beijing in fall 2000 and the Inter-national Conference on Public Communication of Science and Technologyin Geneva in winter 2001, both of which were reported in this journal (seeScience Communication, vol. 22, no. 4, pp. 438-441 and pp. 442-444); andspecial reports and books from myriad organizations and individual authors.From these studies, we have learned a lot about the public communication ofscience and technology—and we have learned that there are also significantgaps in our understanding of this multifaceted process.

Most of this special issue of Science Communication is devoted to taking alook at this robust research area. The issue developed out of a special projectthat began three years ago with funding from NASA’s George C. MarshallSpace Flight Center in Huntsville, Alabama. As Rick E. Borchelt, whochaired that effort, explains in his report of that undertaking, the project—dubbed R2 for Research Roadmap—was designed to examine scholarlyresearch in the public communication of science and technology, suggestareas where additional research was needed to answer some of the key ques-tions that were still unanswered, and identify best practices in the public

Science Communication, Vol. 23 No. 2, December 2001 95-96© 2001 Sage Publications

95

communication of science and technology from around the globe. Towardthat end, the project commissioned a literature review and several researchinitiatives and organized a “best practices” conference for March 2002.

The results of two of those research initiatives, which address specificgaps in our understanding of audiences, are reported in this issue. The first ofthese, by Susanna Hornig Priest of Texas A&M University, examines the roleof trust in judgments people make about the riskiness of new technologies,such as biotechnology. The other, by Mark Tremayne and Sharon Dunwoodyof the University of Wisconsin–Madison, advances our understanding ofhow visitors interact with scientific Web sites and points us in the direction offuture studies that can illuminate this rapidly developing area. Earlier ver-sions of both of these two articles were presented at the annual meeting of theAmerican Association for the Advancement of Science in February 2001 inSan Francisco.

This special issue continues with an article by Robert A. Logan of the Uni-versity of Missouri, who provides a conceptual history of the field, tracing itsorigins through the lens of the scientific literacy tradition and the interactivescience model. This article is followed by the literature review commissionedby the R2 project and undertaken by Michael F. Weigold of the University ofFlorida. The literature review organizes this diverse field by focusing on thekey players in the science communication process, including scientists them-selves, as well as news organizations, journalists, science information profes-sionals, and audiences.

The special issue concludes with the project report by Rick E. Borchelt,who outlines the project’s findings and recommendations.

Public communication of science and technology is only one aspect of thescience communication process that Science Communication addresses, ofcourse. But we hope that this special issue helps to shed light on the develop-ment of the field and to stimulate our thinking about some of the most excit-ing areas of exploration within it. As one of the participants in the R2 project,I especially want to recognize the contribution of Rick Borchelt, whosevision and commitment are reflected in both the project itself and this specialissue.

—Carol L. RogersEditor

96 SCIENCE COMMUNICATION

SCIENCE COMMUNICATIONPriest / COMMUNICATION VARIABLESScience communicators and the broader scientific community often expect media informationcampaigns to mold attitudes about science and technology in predictable ways. But resistance totechnology is not always based on ignorance, and the ability of media-based education todirectly shape opinions is actually quite limited. This article uses data from a recent U.S.national survey on opinions about biotechnology to argue that trust in institutional actors is abigger factor than genetic knowledge in predicting encouragement for specific applications ofbiotechnology. The results have implications for the practice of science, as well as for the prac-tice of science communication.

Misplaced FaithCommunication Variables as Predictors of

Encouragement for Biotechnology Development

SUSANNA HORNIG PRIESTTexas A&M University

Conventional wisdom relies on public information officers, public relationspractitioners, and journalists to inform the nonscientist, or lay, public aboutthe benefits (and sometimes the risks) of new technologies. This is some-times indistinguishable from what Nelkin (1995) called, in a slightly differ-ent context, “selling science.” The implicit intent is usually the modificationof resistant antitechnology attitudes through education, especially educationabout the science underlying the technology. The strategy rests on several

Author’s Note: This article is based on a reanalysis of data collected in cooperation with theInternational Research Group on Biotechnology and the Public, coordinated by George Gaskellat the London School of Economics, and on survey questions developed by that group and byEdna Einsiedel at the University of Calgary, Faculty of Communication and Culture. Fundingfor the original data collection was provided by Texas A&M University. This particular analysiswas supported by a grant from the Marshall Space Flight Center, Huntsville, Alabama. GodfreyJin-Kai Li, a doctoral student in the Department of Sociology at Texas A&M, assisted with theproduction of the statistical results. Address correspondence to Susanna Hornig Priest, TexasA&M University, Department of Journalism, College Station, TX 77843-4111; phone:979-845-5396; fax: 979-845-5408; e-mail: [email protected].


97

questionable assumptions, including the ideas that risk is a characteristic thatcan be accurately calculated by experts (Hornig 1993) and that if peoplerespond negatively to science and technology it is primarily because of a“knowledge deficit” (Ziman 1992).

The “knowledge deficit” approach ignores the evidence that risk is par-tially a matter of value judgments (Hornig 1993) that vary according to bothculture and position in the social structure and that in modern societies risksare managed by complex institutions with persistent images and reputations.It also ignores the fact that risks are not usually distributed randomly but thatexposure to risk is also related to social power (Beck 1992). However, itremains tempting for the scientific community and those who speak in publicon its behalf to assume that dissent generally represents ignorance and that itcan therefore be reduced or eliminated by education, rather than to assumethat dissent represents preference, position, or responses to authoritativeinstitutions.

Where resistance is not attributed to ignorance, it is often attributed toexaggerated or amplified fears induced by too much media attention (e.g., seeLeahy and Mazur 1980). This is a convenient explanation for the failure ofmedia content to predict public opinion, but it is equally incomplete.

Biotechnology is an intriguing example of public understanding of riskbecause of its novel and complex character. Biotech is more controversial inthe United States than has generally been assumed (Priest 2000). Nearlyone-third of the U.S. population believes that genetic engineering is likely tomake the quality of life worse in the future, nearly the same proportion asbelieve that nuclear power generation is likely to do so (although there is noevidence that these are generally the same people).1 A “spiral of silence”(Noelle-Neumann 1984) seems to have developed for this issue for a host ofreasons that include the structural characteristics of U.S. media systems, themedia’s dependence on biotechnology’s promoters as sources, and a climateof opinion that has tended to marginalize criticisms (Priest 2001).

Thus, controversy and objections arising in other parts of the world, rang-ing from India to continental Europe, seem to have come as a great surprise toU.S. opinion leaders and policymakers, as did the extent of objections withinthe United States to a U.S. Department of Agriculture proposal to allow foodslabeled “organic” to be bioengineered. Food processing companies and foodretailers alike scramble to keep ahead of public opinion, and medicalresearchers worry that public opinion will turn against cutting-edge geneticresearch.

None of this is to suggest that the average lay individual understands all ofthe science behind biotechnology. But the dominant assumption that objec-tions to biotechnology necessarily spring from ignorance is itself mis-


informed. Recent data from the National Science Foundation suggest thatobjections to biotechnology are becoming slightly more prevalent among themost educated segment of the population (National Science Board 2000). Forthe U.S. population as a whole, perceived morality appears very closelyrelated to perceived acceptability of various applications of biotechnology,both agricultural and medical (Priest 2000). This is entirely consistent withearlier research that showed a range of considerations, including ethical,environmental, regulatory, communication, and socioeconomic dimensions,enter into people’s judgments of the riskiness of biotechnology, as they do forother technologies (Priest 1995).

This article reports on an analysis of data from a public opinion surveyconducted in April and May of 2000 with 1,002 individuals across the UnitedStates.2 The particular analysis presented here is not concerned, however,with the whole universe of factors that might explain public opinion forma-tion in this area but only with a few of those that might be most directlyrelated to science communication activities: level of knowledge of relevantscience, level of concern with other food safety issues (used as a proxy forgeneralized risk aversion),3 level of trust in particular scientific institutions,and level of awareness of biotechnology.

While it is important to stress that this is a post hoc, exploratory analysis ofa study not specifically designed to test hypotheses about communicationvariables, the results allow us to put the influence of genetics-related knowl-edge into a comparative context and to make a rough judgment about its rela-tive contribution to support for biotechnology. The results, although notdefinitive, certainly direct our attention to the importance of trust in relevantsocial institutions as a predictor of encouragement of these technologies,measured as the total number of applications out of a set of six specific exam-ples provided in the survey that each respondent felt should be encouraged.4

Level of genetic knowledge was also a predictor of encouragement, but therelationship is not as strong as it is for institutional trust.

This project also explored the influence of demographic factors and foundthat age, gender, household income, general level of education, and whether arespondent recalled taking six or more science courses in college also wereimportant in explaining how opinions in this area seem to be formed. Olderpeople are more cautious, as are women; having higher income, havinghigher education, and taking science in college were associated with higherlevels of support.5 Somewhat surprisingly, neither political affiliation nordegree of religiosity seemed to be important predictors of these opinions.This may be because objections come from a variety of groups present in thepluralistic political culture characteristic of the United States. These groupscould conceivably range from religious fundamentalists, who tend to be on

Priest / COMMUNICATION VARIABLES 99

the political Right, to environmental activists, who tend to be on the politicalLeft. But the data did not reveal simple or clear relationships between thesetwo variables and degree of encouragement, so political affiliation anddegree of religiosity are not considered further in this study.

Method and Results

A technique called path analysis and a data set generated by the recentU.S. national survey of public attitudes toward biotechnology (describedabove) (Priest 2000) were used to develop a model to test the conventional“knowledge deficit” and “risk aversion” explanations against a competingexplanation in terms of trust in relevant institutions. The respective contribu-tions to beliefs about whether biotechnology should be encouraged wereassessed for the following factors: general concerns about the food supply,awareness of media messages, education, scientific education, and geneticknowledge, along with institutional trust. Our results, based on a surveydesigned for somewhat different purposes, should be considered exploratoryand not proof that our theoretical reasoning is correct. But, while furtherresearch in this area is called for, we can at least claim that our results stronglysuggest that explanations of public attitudes toward science in terms of eitherscience literacy or the short-term impact of science journalism (at least forbiotechnology) are incomplete. Especially in this particular case, in whichfaith in government regulators and other watchdogs is limited, confidencethat those promoting this kind of science and technology are doing a good jobfor society is more important.

The path analytic approach proved useful to investigate this problem. Thepath analysis was based on regression equations used to calculate coefficientsrepresenting the strength of relationships among the multiple intercorrelatedindependent and mediating variables that were of interest and a dependentvariable, encouragement for biotechnology development. Definitions ofvariables to include as independent, mediating, and dependent factors aresomewhat arbitrary in this kind of model building and must be theoreticallydriven. The details of the model that results, including the values of the corre-sponding path coefficients, are naturally dependent on the variables includedand the means used to measure them. Results need to be interpreted with cau-tion; path analysis may or may not allow researchers to choose among com-peting models of relationships. Nevertheless, path analysis is very useful forvisualizing the relative contributions of various factors to a particular out-come, such as the relative contributions of demographic and attitudinal fac-tors on voting behaviors or the relative contributions of familial and edu-


cational influences on career success. The strength of the associations amongsuch a complex variable set are indicated by the magnitude of the path coeffi-cients (standardized beta coefficients from the regression analysis).

In this type of analysis, independent variables are typically defined asthose demographic factors that are believed relevant, since demographic fac-tors logically precede (and influence) message interpretation and attitudeformation. Independent variables as defined in this model found to predictencouragement for biotechnology, as stated above, were gender, age,income, education, and whether a respondent recalled taking six or more sci-ence courses in college. The latter variable was designed to identify respon-dents with an unusually high degree of interest in science, although that vari-able was not measured directly, under the assumption that those who took sixor more college-level science courses would consist of science majors orminors and others with a particular interest in science. Neither political norreligious affiliations were included, as exploratory analysis did not indicateany significant relationship between these factors and support for biotechnol-ogy. (For our purposes, relationships meeting a criterion of statistical signifi-cance at the .05 level are reflected in our model.) While education levels andcollege study of science may themselves be predicted by income, age, andgender to some degree, we viewed them as logically existing prior to bothexposure to messages and the formation of attitudes regarding biotechnology.

As indicated earlier, mediating variables chosen were levels of generalfood safety concern, awareness of exposure to biotechnology news, trust inrelevant institutions, and level of genetic knowledge. Food safety concernwas assessed using an index based on a series of questions about concern overpesticides, nutritional value, food-borne diseases, and other related issues.This variable was included to test the idea that those with a high general levelof concern about food safety issues might be the same individuals expressingconcern about biotechnology, including food biotechnology. We consideredit a reasonable indication of general risk aversion; ideally, a comparable indi-cator more relevant to medical applications also would have been used.Awareness of biotechnology issues was assessed by an index composed ofanswers to three questions about frequency of conversation, length of timesince the respondent had first heard about biotechnology, and self-perceptionof how well-informed the respondent is.6

Trust in relevant institutions was assessed using an index composed ofanswers to specific questions about whether these institutions were “doing agood job” for society with respect to biotechnology. Level of genetic knowl-edge was assessed using test questions developed by the parallel Europeanand Canadian research projects (for initial findings from Europe, see Gaskellet al. 2000; for Canada, see Einsiedel 2000). These consisted of a series of ten


true-and-false questions ranging from relatively simple (whether only genet-ically engineered tomatoes contained genes) to more complex (whetherhuman and chimpanzee DNA are closely related).

Other trust (“doing a good job”) items had been included in the originalsurvey, but only four were incorporated into this model. As Table 1 indicates,positive impressions of scientists, industry, government regulators, and con-sumer organizations most strongly predicted encouragement of biotechnol-ogy applications when these were considered as individual items. However,factor analysis had been performed, suggesting that faith in four sets ofinterests—science/scientists, farmers/agriculture, food retailers, and the bio-technology industry—were highly intercorrelated, suggesting they formed asingle attitudinal dimension. This factor, not other factors involving impres-sions of various watchdog or oversight groups (ethics committees, the gov-ernment, the media, consumer groups, and churches), proved in turn to be thebest predictor of the encouragement of biotechnology variable when the fac-tors were considered as composites. Only the “key institutional providers”data were therefore included in this model (scored as the sum of answers tothe four individual questions about science, industry, farmers, and stores), aswe felt that incorporating the individual items separately would complicatethe model without providing good theoretical grounds on which to suggest auseful interpretation.

Table 1 also suggests, independent of the other factors studied, that faith inregulatory agencies (i.e., belief that they are doing a good job for society inthe area of biotechnology) is not high, nor is faith in media. Faith in scientificinstitutions remains extremely high. Faith in industry, while high, is lower.

The dependent variable was a composite of encouragement items for eachof six forms of biotechnology. Respondents were asked whether each of thefollowing six applications should, on the whole, be encouraged (or stronglyencouraged): cloning animals, genetic testing for disease, food biotechnol-ogy generally, genetic engineering of crops, development of animal organsfor xenotransplantation to humans, or engineering bacteria to producepharmaceuticals. (These questions were also based on those used in the 1999Eurobarometer and the parallel Canadian telephone survey conducted in2000.) Either positive response (encouragement or strong encouragement)was coded as 1. Thus, theoretically, a given respondent’s overall level ofencouragement score could range from 0 to 6.

The results are summarized in Figure 1, which shows the relative strengthsof relationships (expressed as partial correlation coefficients) among thevariables of interest. Clearly, while other food safety concerns, knowledge ofgenetics, and awareness of related news are all weakly related to encourage-ment of biotechnology, it is trust in the four key institutions of science,


agriculture, food retailers, and biotechnology corporations that best predictsencouragement of the further development of these applications.

To satisfy ourselves that the influence of genetic knowledge on encour-agement had not been obscured by treating all six applications together in ourmodel, we also calculated Pearson correlation coefficients between this scoreand level of encouragement for each of the six applications taken separately(see Table 2). Some variation in this relationship is apparent, with coeffi-cients ranging from .084 for the genetic testing application to .232 for thefood biotechnology application. All correlations were positive and statisti-cally significant at the .05 level or better; however, these are not strong


TABLE 1Belief in Whether Various Institutions Are Doing a Good Job

for Society with Respect to Biotechnology DevelopmentRelated to Average Number of Biotechnology Applications

Encouraged for 1,002 U.S. Respondents

Encouragement EncouragementScore for Score for

Respondents RespondentsValue for T,

Agreeinga DisagreeingGroups

Institution M n M n Comparison Significance

Media 4.21 445 4.01 444 1.682 .093Industryb 4.44 580 3.28 225 7.919 .000Ethics committees 4.18 472 3.95 308 1.720 .086Consumer

organizations 4.20 591 3.62 213 4.078 .000Environmental

groups 4.08 515 4.07 283 0.071 .943Government 4.34 396 3.80 432 4.363 .000Shopsb 4.16 586 3.83 333 2.621 .009Farmersb 4.09 729 3.72 141 2.097 .037Churches 3.96 369 4.17 409 –1.632 .103Scientistsb 4.32 779 2.45 92 9.383 .000Doctors 4.18 586 3.76 282 3.285 .001

NOTE: Considered as independent variables, neither trust in industry, in consumer groups, ingovernment, nor in doctors was highly predictive of encouragement. Treated as an aggregatevariable based on factor analysis results, trust in the combination of industry, shops (stores),farmers, and scientists was the most predictive of encouragement.a. That is, for respondents agreeing that the particular institution is doing a good job. Encourage-ment score is the total number of applications, out of six, the development of which each respon-dent agrees or strongly agrees should be encouraged.b. Based on the results of a factor analysis, responses for these institutions were grouped to-gether; a variable composed of the agreement variable (trust) for all four institutions was the pre-dictor of encouragement used in further analysis.


Food Concern

Index

College Science

Income

Education

Age

Gender

KnowledgeScore

AwarenessIndex

Trust in Biotech Providers Encouragement

Score (Composite forSix Applications)

-.164 -.109

.273

-.087

.220

.255

-.080

.082

.164

.095

-.131

.281

.084

.169

-.081

Figure 1: Path Diagram Showing Relative Strength of Relationships among Various Pre-dictor Variables for Encouragement of Biotechnology (Six Specific Applica-tions Combined)

Trust item includes responses for only four key institutions (industry, scientists, farmers, andshops) that were shown to be interrelated on the basis of factor analysis results and that collec-tively predicted encouragement; other institutional trust items were also individually significant.

TABLE 2Mean Degree of Encouragement and Correlation Coefficients betweenScience Literacy (Knowledge) Scores and Degree of Encouragement

(Four-Point Scale from Definitely Disagree to Definitely Agree)for Six Applications of Biotechnology (N = 1,002)

Correlation (Pearson’s r)Application Encouragement (M) with Knowledge Score

Cloning 2.54 .195**Xenotransplantation 2.64 .125**Food biotechnology 2.75 .232**Plant engineering 3.03 .186**Bacteria engineering 3.28 .130**Genetic testing 3.35 .084*

*Correlation is significant at the .05 level (two-tailed test). **Correlation is significant at the .01level (two-tailed test).

relationships. (For the strongest, that for food biotech, a correlation of thismagnitude indicates that only about 5 percent of the variation in encourage-ment can be predicted—or, in the statistical sense, explained—by variationsin knowledge.)

Discussion

While those in the scientific community complain that media stories aretoo often inaccurate, incomplete, or misleading, media scholars interested inscience communication typically offer a slightly different sort of criticism.This latter group is more likely to be concerned with accounts that focus toocompletely on breakthroughs, suggesting a distorted vision of the process ofscience; that are too dependent on the “information subsidies” provided byuniversity public affairs departments, government information officers, andcorporate public relations staffers; or that consistently fail to recognize theexistence of intellectual controversy and of complex ethical and policy con-siderations. What is at stake from this point of view is not so much the accu-racy of information vis-à-vis a specific technical development or scientificissue but the contribution to a general understanding of the nature, currentstatus, and social significance of the scientific endeavor and its results. Oneconsequence is that when controversies of this type inevitably arise, as withissues ranging from global warming to human cloning, neither journalists northe public have much context for understanding them. Either social inertia oroverreaction conceivably may result. Productive democratic dialogue, how-ever, is likely to be thwarted. Public understanding of the nature and role ofkey institutions is part of the necessary context.

Commonly, from either perspective, science journalism is assumed to bean important determinant of science-related attitudes and opinions, primarilythrough its explanations of scientific results. Media research relevant to otherquestions about effects suggests that if this is true, the important influences ofscience journalism are long term and indirect. Of course, general scienceeducation is also important. But, for those who are not scientists or universitystudents, media accounts are likely to be very important sources of informa-tion on new developments in science. Still, their main influence is unlikely tobe found in direct short-term impact on attitudes or opinions. Over the longhaul, the cumulative impressions left by these accounts about the nature ofscience, of scientists, and of the organizations and agencies that sponsor andmake use of scientific results are undoubtedly more influential.

This statement captures processes that media researchers refer to as “culti-vation” effects or the “social construction of reality.” Mediated information


helps shape our ideas about how the world works and who is to be trusted.Media accounts can also (in concert with other social institutions and pro-cesses) help define which issues will garner the most public attention (agendabuilding) and, arguably, serve to define the important aspects or the funda-mental natures of these issues in specific ways (framing). But the media arenot the controlling influence on the collective public response. In the case ofthe particular set of issues investigated in this analysis, those surroundingbiotechnology, it is the long-term cultivation of a particular view of scientificinstitutions that is probably more important than the short-term, perhaps tran-sient, effects on public opinion of particular media accounts. Such a viewmay see science as serving broader social interests or narrow industrial inter-ests and as being benign or threatening, and it may see technology as improv-ing or eroding the quality of life.

The public images of science, scientists, and scientific institutions that areconstructed by both media accounts and exposure to science education nodoubt contribute, in turn, to the attitudes and opinions that form the context inwhich future news of scientific events is understood. A public that under-stands nutrition research (for example) as a series of definitive experimentswill be mystified when they read of results suggesting contradictory dietaryguidelines. A public that believes scientific results to be fixed, static, and cer-tain may be confused by an ongoing series of revisions to evolutionary the-ory. But it does not necessarily follow that the public can easily be turnedagainst a particular scientific or technological development because of minortechnical inaccuracies or “wrong” choices of terminology on the part of sci-entific journalists. Nor can public enthusiasm be guaranteed by the rightpress release. As the scientific community becomes ever more dependent onprivate industry for funding in the face of the shrinking federal dollar forbasic research, it might well be concerned with whether its public personastands to be affected by the association.

Public resistance to genetically engineered foods cannot reasonably beentirely attributed to tabloid descriptions of them as “frankenfoods” or poorexplanations of how recombinant DNA actually works, and their mixedreception cannot be cured by a heavy dose of revised technical information.In fact, judgments about the riskiness of new technologies such asbioengineered foods involves judgments about the trustworthiness of scien-tists and their employers—and in this particular case, the trustworthiness ofother groups such as farmers, grocery retailers, and corporate leaders aswell—not just judgments about the science itself. And, like our understand-ing of science, our trust in these various entities and individuals is not theshort-term result of individual news stories but is gradually built up over time,with experience.


In other words, faith in the power of news accounts to generate (short term,at least) either public support or public opposition to particular forms of sci-ence and technology is misplaced. News stories do not have this sort of magicbullet effect. Public trust is a valuable commodity, and it appears to be some-thing much more stable and difficult to manufacture than is generallyassumed. Often, this is perceived as a challenge to those who seek to shapepublic opinion: institutional reputations are difficult to alter, and since publicresponse to science and technology depends in part on how relevant institu-tions are perceived, it will be concomitantly difficult to control. But there iscertainly a bright side to these dynamics. Few of us would want to live in aworld in which the reputations of institutions were too easily manipulated bymedia messages.

Recognition of the relationship between trust in relevant institutions andacceptance of particular technological applications—such as biotechnology—should not encourage our redefinition of the science public relations problemas a question of raising levels of trust in a strategic way. Of course, high levelsof public trust and confidence would be a good thing from the point of view ofscience. But blind public faith in any institutions, even the institutions of sci-ence, is neither healthy nor sustainable. Policy decisions remain matters fordemocratic debate, not scientific determination. Both science journalism andscience education should seek to create citizens who will remain among theinterested spectators of science but who are also not afraid to assume respon-sibility for considering its policy dimensions. Some commentators have sug-gested that science (or “scientific”) literacy be redefined to include under-standing of how science is done, not just its results (Maienschein 1999). Italso needs to include an understanding that the boundary between scienceand policy is dynamic, that the impact of science is often uncertain, that sci-ence and technology policy reflects value-based decision making, and thatthe equitable distribution of risks and benefits associated with science andtechnology remains a substantial challenge.

Similarly, public mistrust has deep roots; it is a mistake to assume thatpublic resistance to technology is always a function of misinformation ortransient overreaction. The collective memory of contemporary society isfull of examples of technological (often environmental) disaster. This is thebedrock of public attitudes on which opinions about new technologies mustrest. Fortunately for technology’s promoters, this bedrock appears underlainby a yet deeper layer consisting of cultural faith in the great power of both sci-ence and technology for doing good. Levels of scientific knowledge and edu-cation are no doubt related in complex (largely indirect) ways to some ofthese attitudes, but they do not fully determine opinions, especially for con-troversial areas of science policy and ethics.


Our study involved communication-related and demographic variablesonly. Limited by the nature of the data set, we did not attempt to measure orassess values directly, although these are very likely to be related in turn to thedemographic factors (gender, age, income, education, college study of sci-ence) that we did include. And variations by income and education mightwell be reflective of broader class distinctions that we also did not assessmore directly, while the statistical influence shown here of gender almost cer-tainly reflects both value differences and persistent gender-based differentia-tion of social roles. In short, we do not claim that this study reflects or incor-porates all or even most of the variables of interest and importance in thecontext of an exploration of social aspects of risk response. But we do claimthat our results demonstrate that explanations based on knowledge deficits oraversion hypotheses alone are both incomplete and misleading. The dynam-ics of the social contracts between key institutions and the public are also inplay. Work continues to explore how these issues play out across national andcultural boundaries.

Where does all this leave science communication? It is not a magic bulletresource for producing public acceptance of science and technology. Sciencecommunication has a crucial role to play in the generation and maintenanceof public trust in science, but not at the expense of ignoring the imperfect cer-tainty of scientific conclusions, the value- and power-laden character of thesocial uses of science, or the nature of the social contract between the scien-tific community and the broader public. Rather, science journalism has awatchdog role to play for science and technology policy, just as it does forother aspects of the political scene.

The bottom-line recommendation of this research is not a call for intensi-fied institutional promotion—trust is important, but it cannot be marketedlike soap. The social contract underlying public support of science is a fragileone. Education, even science education, will not guarantee that that supportwill be continued. Only sustained attention to the generation and mainte-nance of public trust through stressing the public service obligation of the sci-entific community, a role that could be threatened by increasing commercial-ization, will hold the promise of providing such a guarantee.

Notes

1. About 42 percent of those who are pessimistic about nuclear energy are also pessimisticabout genetic engineering, whereas about 30 percent of the general population is pessimisticabout genetic engineering.

2. The telephone survey was conducted, using standard random-digit dialing techniques, bythe Public Policy Research Institute at Texas A&M University. Early U.S. results from this sur-


vey were published in Priest (2000). Other results of the survey, in particular the comparison ofthe U.S. results with those obtained in Canada and in a number of European countries throughthe Eurobarometer project and supplemental efforts in additional countries, are forthcoming in abook to be published by the London Science Museum.

3. I am not unaware of the commonly noted differences between public perception of agri-cultural and of medical biotechnologies. There is some suggestion in the data from this surveythat this distinction may be more complex than is usually assumed, involving not just the type ofapplication but the type of DNA that might be manipulated; again, see Priest (2000) for discus-sion. For the sake of simplicity, however, the six applications have been collapsed into a singlemodel for purposes of this particular analysis, except where noted.

4. The score for each individual respondent indicates the number of applications, out of sixpossible, the respondent definitely agreed or tended to agree should be encouraged.

5. Independent of the effects of the mediating variables, women were also less encouragingof these technologies than were men, older individuals were less encouraging of them than wereyounger ones, and higher-income individuals were more encouraging of them than werelower-income individuals, even after all other factors we included in our analysis were con-trolled. Although these factors do not seem directly related to communication issues and are notemphasized in this discussion, these differences seem potentially important because they arelikely to reflect value orientations. Women are traditionally responsible for both the health offamilies and the food they eat. Higher-income individuals may be differentially attracted to theinvestment opportunities or simply the technical novelty of these technologies. And, it is credi-ble that age might also be a factor predictive of responses to technological innovation, but again,not one directly relevant to this study of possible communication influences.

6. The questions included were as follows: “Before today, have you ever talked about modernbiotechnology with anyone?” (rated on a scale from 0 to 4, from never to frequently); “How re-cently do you remember first hearing about modern biotechnology, whether from the news orfrom another person?” (rated on a scale from 1 to 4, with higher numbers indicating a longer timesince first hearing); and “In general, how well informed do you feel that you are with respect tomodern biotechnology?” (rated on a scale from 1 to 5).

References

Beck, U. 1992. Risk society: Towards a new modernity. London: Sage.Einsiedel, E. 2000. Cloning and its discontents—A Canadian perspective. Nature Biotechnology

18 (9): 943-44.Gaskell, G., et al. 2000. Biotechnology and the European public. Nature Biotechnology 18 (9):

935-38.Hornig, S. 1993. Reading risk: Public response to print media accounts of technological risk.

Public Understanding of Science 2 (2): 95-109.Leahy, P. J., and A. Mazur. 1980. The rise and fall of public opposition in specific social move-

ments. Social Studies of Science 10 (3): 259-84.Maienschein, J. 1999. To the future—Arguments for scientific literacy (Commentary). Science

Communication 21:75-87.National Science Board. 2000. Science and engineering indicators—2000. NSB-00-1.

Arlington, VA: National Science Foundation.Nelkin, D. 1995. Selling science: How the press covers science and technology. 2d Ed. New

York: Freeman.


Noelle-Neumann, E. 1984. The spiral of silence: Public opinion, our social skin. Chicago: Uni-versity of Chicago Press.

Priest, S. H. 1995. Information equity, public understanding of science, and the biotechnologydebate. Journal of Communication 45 (1): 39-54.

. 2000. U.S. public opinion divided over biotechnology? Nature Biotechnology 18 (9):939-42.

. 2001. A grain of truth: The media, the public, and biotechnology. Lanham, MD:Rowman & Littlefield.

Ziman, J. 1992. Not knowing, needing to know, and wanting to know. In When science meets thepublic, edited by B. V. Lewenstein, 13-20. Washington, DC: American Association for theAdvancement of Science.

SUSANNA HORNIG PRIEST is an associate professor of journalism at Texas A&M Uni-versity, where she teaches courses on media, science, and society. Her research is onmedia representations and public responses associated with scientific and technologicalrisks and controversies, especially biotechnology. She has recently written a book titledA Grain of Truth: The Media, the Public, and Biotechnology (Rowman & Littlefield2001) on the formation of U.S. public opinion in this area.


SCIENCE COMMUNICATIONTremayne, Dunwoody / WORLD WIDE WEBThis study examines the role of interactivity in the presentation of science news on the WorldWide Web. The authors propose and test a model of interactive information processing that sug-gests a relationship between interactivity, cognitive elaboration, and learning. A think-aloudmethod was employed to provide insight into study participants’ mental processes during Webuse.

Interactivity, Information Processing, andLearning on the World Wide Web

MARK TREMAYNESHARON DUNWOODYUniversity of Wisconsin–Madison

The Web is fast becoming a dominant means of mass communication. Unlikemost older media channels, the Web can be configured to allow synergybetween sender and receiver. Web users, by following links, using searchengines, and selecting items from pull-down menus, have the capacity to takea more active role in information consumption. And, appropriate softwarenow makes it possible for a user and a site to collectively construct meaning.This give-and-take process afforded by the new medium is described bymany analysts as “interactivity.”

The effects of interactivity on users are the subject of debate. Some see thehigher level of user activity leading to greater involvement by the user and,subsequently, more significant media effects. Others see the simple actions ofWeb users as no more significant than television viewers’ channel-changinghabits or readers’page-turning behavior. In this latter view, interactivity is ofno special relevance.

Authors’Note: This research was supported, in part, by a grant from the Marshall Space FlightCenter, Huntsville, Alabama. Address correspondence to Sharon Dunwoody, Evjue-BascomProfessor and Director, School of Journalism and Mass Communication, University ofWisconsin–Madison, 821 University Avenue, Madison, WI 53706; phone: 608-263-4080; fax:608-262-1361; e-mail: [email protected].


111

In this study, we propose and test a model of interactive information pro-cessing. Specifically, we suggest that characteristics of a user and of a Website both influence the level of interactivity employed by the user.Interactivity, in turn, will influence information processing strategies. And,those strategies will work to influence knowledge acquisition.

Study participants spent time on one of two science Web sites, and theiractivities were recorded. We also asked participants to think aloud while theylooked at Web site content as a means of accessing their thought processes.We hoped to reach some conclusions about how Web site features affect anindividual’s information processing.

Interactivity: Origins and Related Terms

The study of interactivity predates the current fascination with new,so-called interactive technologies and the new forms of communication theseinventions are enabling. Goffman (1967) studied face-to-face interaction andconcluded that it is not something to be analyzed at the level of individualsbut rather involves “the syntactical relations among the acts of different per-sons mutually present to one another” (p. 2). Others have applied this idea ofcopresence to mediated forms of communication and have described what(sometimes) happens there as “interactivity” (Rafaeli 1988). For Goffman,communication was just one of many things that can occur during an interac-tion. But, Rafaeli sees full interactivity as just one of many things that canoccur during communication. Some communication, under this conceptual-ization, is not interactive. A message received but not replied to is one-wayand noninteractive. Some response, or feedback, is required to achieveinteractivity. Indeed “feedback,” in the Westley and MacLean (1957) sense ofthe term, is equated by many scholars with “interactivity” (Newhagen,Cordes, and Levy 1995; Rice 1988). Newhagen (1997) suggested using theterm “cybernetic feedback,” which is borrowed from early work in engineering.

Other terms have been used to describe the same, or similar, processes.Bordewijk and van Kaam (1986) proposed several models of “tele-informationservices.” The two that most closely resemble interactivity are their “consul-tation” and “registration” models. In the former, an information companyholds data that are accessed (or consulted) by those who want them; in the lat-ter, individuals possess the data that the information company then seeks out.The authors do not use the word “interactivity,” preferring instead the term“feedback.”


Ball-Rokeach and Reardon (1988) proposed the term “telelogic” todescribe the process of talking or writing while separated geographically.They use both feedback and interactivity as dimensions of telelogic. Thisframework has been applied to Internet research (Ogan 1993). Similarly,Steuer (1992) used interactivity as one dimension of “telepresence,” a con-cept used to characterize the “realness” of virtual realities.

However, “interactivity” is the term most often used by scholars todescribe the two-way nature of most new media technologies (Ha and James1998; Hawkins and Pingree 1997; Kipper 1991; McMillan 1999; Rafaeli1988; Salomon 1990; Sundar, Brown, and Kalyanaraman 1999). Scholarsagree that interactivity is a defining component of new media technologies,but they have not yet reached agreement on how it is best conceptualized.

Interactivity: Conceptualizations and Measures

Is interactivity just feedback? In mass communication models, feedbackis typically illustrated by a dotted line from the receiver to the sender. Thisimplies that the feedback signal is somehow weaker or less important than theprimary signal (which moves from sender to receiver). It also implies that itcomes after the original message sent by the mass communicator. Further-more, it comes back not along the same channel as the primary signal butthrough some other channel. Typical forms of feedback are indirect (circula-tion or ratings data) or infrequent (letters to the editor, calls to the station,etc.).

Does this model adequately reflect communication that takes place on theInternet—the channel most frequently described as interactive? There, mes-sages sent by individuals often precede those sent by mass communicators(by use of a search engine, for example). These messages are not infrequentand are certainly more direct. The signals also travel through the same chan-nel as that used by the mass communicators. So “feedback” in the traditionalsense is not an adequate term to describe the style of communication occur-ring online and elsewhere.

Where Does Interactivity Reside?

Researchers pursuing human-computer interaction have placed emphasison different components of the exchange. These approaches can be groupedin three categories (McMillan 1999). One approach is to focus on the individ-ual and to determine if his or her behaviors are interactive and what effects

Tremayne, Dunwoody / WORLD WIDE WEB 113

such interactivity might have (Hawkins and Pingree 1997; Salomon 1990;Walther 1994). For example, Salomon (1990) focused on the cognitiveeffects of computer use on school children.

Another approach is to focus on characteristics of communication chan-nels and to draw distinctions between new and old media (Kipper 1991;Newhagen 1997; Rice 1988; Rice and Williams 1984; Steuer 1992; Wil-liams, Rice, and Rogers 1988). This is a structural approach to interactivity.Steuer (1992), for example, devised an interactivity scale with print media onthe low end and electronic media such as computer games on the high end.

The third perspective is that offered by Rafaeli (1988). He dismissed thestructuralist approach: “interactivity is not a characteristic of the medium. Itis a process-related construct about communication” (Rafaeli and Sudweeks1997). He proposed that an exchange of communication is interactive only ifeach party is responding to the other in a meaningful way. Rafaeli (1988)defined interactivity as “an expression of the extent that in a given series ofcommunication exchanges, any third (or later) transmission (or message) isrelated to the degree to which previous exchanges referred to even earliertransmissions” (p. 111). This conceptual definition has been used in a num-ber of studies (Newhagen, Cordes, and Levy 1995; Rafaeli and Sudweeks1997; Sundar, Brown, and Kalyanaraman 1999).

These different approaches to studying interactivity produce correspond-ingly different ways of operationalizing it. A simple diagram illustrates thispoint (see Figure 1).

In human-computer interaction (perhaps the most common type ofinteractivity currently under study), the interactivity in the diagram isdepicted by the arrows. These are the messages or signals sent back and forthbetween the agents. But, this is not what most researchers measure. Instead,researchers focus on the H (studying individual users), the C (the structuralistapproach referred to above), or both.

The focus on users or structures is a practical consideration. To focus oninteractivity as a communicative process (as Rafaeli, 1988, advocated)requires access to the messages sent by both (or all) parties. This is easilydone where such a record already exists, as it does for online discussiongroups (Rafaeli and Sudweeks 1997). But for much computer-mediatedcommunication, such a complete record of messages sent and received ishard to isolate. In many cases, information from users can be difficult toobtain for proprietary reasons. So, researchers have focused their attention oneither end of the communication process using (in most cases) an implicitrather than explicit conceptualization of interactivity.


Our Conceptualization of Interactivity

Most researchers use the words “two-way” when describing interactivecommunication. We believe at least two communicators must be involved,each operating as an intentional sender and receiver. This is in line withRafaeli’s (1988) conceptualization, which requires related messages to beexchanged and built on by two (or more) parties. Furthermore, reactive(Rafaeli 1988) communication that involves only one response to a messageis not deemed fully interactive because the second party sees no response tohis or her message. Such reactive communication actually typifies traditionalmass communication where the consumer’s message or feedback often con-sists only of purchasing the paper or subscribing to the cable channel. Theexchanged messages must be related and responsive to each other (for exam-ple, talking with rather than at someone).

We believe these few criteria define “interactivity.” Some researchers alsobelieve that the speed of the exchange is a dimension of interactivity as well.While Rafaeli and Sudweeks (1997) described interactivity as “simulta-neous” message exchange, there is little reason to believe that asynchronousexchanges (such as e-mail or voice mail) are not interactive. We see this as anattribute of media channels rather than a dimension of interactivity itself.

Interactivity Operationalized

This study compares the use of two science-oriented Web sites. The par-ticipants in interactivity are, therefore, individual users on one side, and theWeb sites (and the organizations behind them) on the other. When a userarrives at a Web site and begins reading (or “viewing”), an initial messagefrom the Web publisher has been sent and received. The user may then chooseto select a link or use a search engine to obtain additional information. This ismessage-sending activity on the part of the user. If the link or search engineworks, then the request will be received by the Web server and a new messagesent in response. These three messages (two by the Web site and one by theuser) form the necessary trio required for interactivity. By recording each ofthe interactive instances, an interactivity score can be calculated for thesession.


Figure 1: A Typical Interactive ProcessNOTE: H = human; C = computer.

Message-sending activity by a user (using either a keyboard or a mouse)that results in a corresponding (and responsive) change in screen content is,therefore, our operationalization of interactivity.

Web Site Structure

Rafaeli and Sudweeks (1997) insisted, and we agree, that communicationcan be more or less interactive, and the variable is therefore continuous.While it may be appropriate to describe one medium as more interactive thananother, the evidence for this should be found in the proportion of messagesending being done by each party rather than in the structural characteristicsof the channels. A medium that appears to support interactive communica-tion may nevertheless be used in much the same way as a unidirectionalmedium.

The structural characteristics of any medium, however, do place limits onthe type of communicative exchanges that are possible there. Steuer (1992)defined interactive media as those that allow users to modify the mediatedenvironment and include speed and range among their dimensions. In thisview, media that allow users many opportunities (range) to alter contentquickly are the most interactive.

This study involves a comparison of two publications presented on thesame channel (the World Wide Web), so there will be no significant differ-ences in the speed of interactions. The range, or the amount of interactiveoptions available to users, will be varied to examine the impact of range oninteractivity.

We operationalize interactive Web site structure as the degree to whichusers are offered choices that allow them to alter the content they consume.These options include hypertext links, search engines, and rollover graphics.

Web Experience

Another precursor to interactive behavior resides in the characteristics ofusers. Many individual traits could affect a person’s interest in interactivecommunication. Among these, and of special interest for this study, is experi-ence with the medium, in this case the Web. Researchers have identified a linkbetween a person’s previous experience with an activity and his or her currentefficacy level toward that behavior (Bandura 1994). As one’s comfort levelwith a behavior rises, so does the likelihood of engaging in that behavior in


the future. We operationalize Web experience as the amount of self-reportedWeb use in a typical week (for either personal or work-related reasons).

Cognition and the Web

Communications researchers studying interactivity have focused eitheron the structural features of channels allowing two-way discourse or on theactions of those using these structures. So far, relatively little attention hasbeen paid to the cognitive activities of those engaged in interactive behavior.One purpose of this study is to identify the types of cognitive informationprocessing activated during interactive media use.

Cognitive science is certainly a natural fit for study in this area. The field ishistorically interdisciplinary and uses the computer as a metaphor for under-standing the workings of the human mind (Gardner 1985). Human percep-tions, such as vision, have been described by cognitive scientists as interac-tive processes (Gibson 1966; Shaw and Bransford 1977).

The human-computer metaphor has been extended to information pro-cessing (Simon and Siklossy 1972). Under this conceptualization, an individ-ual takes in information through the five senses and stores it in temporary (or“working”) memory before transferring some of it to long-term memory andstoring it in an associative network, sometimes referred to as “schemata”(Rumelhart, Lindsay, and Norman 1972; Wicks 1992).

Rehearsal

Individuals employ a number of strategies to embed new information inmemory. One such strategy is rehearsal (or repetition of information), alsoreferred to as cognitive maintenance. The use of flash cards to remembermultiplication tables is a typical rehearsal strategy. While not necessarily thebest means of storing information in memory, this strategy has been demon-strated to improve rote recall (Belmont and Butterfield 1971).

In the context of information processing during Web use, cognitiverehearsal is simply the reading of material presented with no effort made toconnect it to related material (from either prior knowledge or material pre-sented simultaneously).

Elaboration

For retention of more complex information, another strategy is consideredespecially effective. That strategy—cognitive elaboration—is of particular


interest in this study. When an individual encounters a new piece of informa-tion, he or she may connect it to related, preexisting knowledge or to otherinformation encountered at the same time. The more connections the individ-ual makes, the more he or she is elaborating on it, and the more strongly heldin memory this new information should be (Anderson 1990; Anderson andReder 1979; Eveland and Dunwoody 2000). Information that an individualconnects to previously held information will become embedded in that per-son’s associative network, while information that is merely noticed will beheld (if at all) in relative isolation. A wide body of literature examines the roleof cognitive elaboration on memory and learning. Reviews of some of thatwork in educational and cognitive psychology (Estes 1988; Greene 1992)conclude that processing information elaboratively improves performanceon memory posttests.

Much of the research on cognitive elaboration has focused on learningfrom text (Baker 1989; Hamilton 1997; McDaniel and Einstein 1989). Nowscholars are beginning to apply these techniques to hypertext (orhypermedia) as it appears on the Web (Eveland and Dunwoody 2000). Mate-rial presented in this fashion allows for “interactive reading” (Burbules andCallister 1996). In this case, “the reader becomes the constructor of his/herunique text” (Lawless and Kulikowich 1996, 386). But how much flexibilitydoes one truly have in a hypertext, or hypermedia, environment? It is up to thedocument’s designer. A nonlinear text, while still finite, can be designed sousers with diverse backgrounds can navigate the text according to their indi-vidual needs and interests (Fredin 1997).

If the user, while reading material presented this way, makes meaningfulconnections between new information and that already held in memory, cog-nitive elaboration is occurring.

(Dis)Orientation

A drawback to hypermedia presentations that was recognized by severalresearchers in the 1990s is disorientation (Foltz 1996; Landow 1991;McDonald and Stevenson 1996; Rouet and Levonen 1996). As users followlinks into a hyperdocument, they can lose track of where they are with respectto the information space. Disorientation has been demonstrated to have aninhibiting effect on learning because effort spent on orienting behavior drawson limited cognitive resources (Marchionini 1988; Mayes, Kibby, andAnderson 1990; Tripp and Roby 1990). This is the so-called cognitive loadeffect.

Previous research on how people process information during hypermediaexposure has revealed orientation to be the most common cognitive activity


(Eveland and Dunwoody 2000). The researchers predicted, however, that thismight change over time. As people become more accustomed to using theWeb and designers improve the navigability of Web sites, the need for cogni-tive orientation may diminish.

Learning from Media

A great deal of research has focused on learning from media and the influ-ence of structural differences on knowledge acquisition as measured by recall(a review appears in Neuman, Just, and Crigler 1992). Because most mediause is not specifically goal oriented, retention of information is usually low,regardless of medium. Experimental and survey research most often findsprint media superior to broadcast media for knowledge acquisition (Furnhamand Gunter 1989; Halpern 1997; Wicks 1992). However, this may not stemfrom structural characteristics of the media. Survey research that measuresonly “time spent with” rather than attention can underestimate knowledgeacquisition from television (DeFleur et al. 1992). And, there is some evi-dence of cognitive differences in the way visual information is processedwhen compared with print that requires a different experimental measure ofknowledge acquisition (Shoemaker, Schooler, and Danielson 1989).

Certain characteristics of individuals may affect their performance onmemory tests. Interest in the subject is a major contributor to recall and com-prehension of stories in the media (Berry 1983; Booth 1970-1971; Woodall,Davis, and Sahin 1983). A related factor is prior knowledge of a subject. Forexample, a person with an extensive background in a subject is more likely tofind a place to connect new information in his or her associative memory net-work than is a person who is new to a topic. Knowledge of current events is agreat predictor of recall and comprehension of new stories in the news (Priceand Zaller 1993; Woodall, Davis, and Sahin 1983).

There are a number of ways to quantify knowledge acquisition. The fourmost common are tests of knowledge using free recall, cued recall, recogni-tion, and comprehension. For free recall, study participants are asked toremember anything they can from a media-use session. Cued recall is used toactivate certain areas of knowledge (who, where, etc.) and gives participantsa place to begin. Recognition involves closed-ended questions and has beensuggested as the most appropriate measure for knowledge gain from broad-cast media (Shoemaker, Schooler, and Danielson 1989). Comprehensioninvolves more than the knowledge of independent facts; rather, it requires anability to integrate those facts into a meaningful system. This study employs acued recall measure of knowledge acquisition.


Theoretical Model and Hypotheses

The conceptual discussion above leads us to the following model of inter-active information processing (see Figure 2).

The model shown in Figure 2 can be considered in two halves labeled theaction phase and the cognition phase. The action phase concerns the actualphysical actions of users. We propose here that the amount of interactivebehavior in which an individual engages is a function of the structural charac-teristics of the medium (in our case Web sites) and individual user traits (suchas Web experience). Specifically, we predict the following:

Hypothesis 1: A Web site with a greater range of user options will be associatedwith higher levels of interactivity than one with a lower range of options; and

Hypothesis 2: Individuals with high levels of Web use will engage in more interac-tive behavior than will individuals with low levels of Web use.

If the action phase hypotheses are correct, experienced Web users on a com-plex Web site will exhibit the highest levels of interactive behavior whileusers with less experience on a site with fewer options will exhibit the lowestlevels of interactive behavior. If these hypotheses are confirmed, we will beable to do a between-groups comparison of the effects of action on cognition.

The cognition phase concerns mental activities: how the user processesinformation and the effects of that cognitive activity on knowledge gain. Wepropose here that two types of cognitive effort—elaboration and rehearsal—will promote knowledge acquisition while another—orientation—will not.


Figure 2: Interactive Information Processing Model

Interactive behavior can be coupled with any of the cognitive informationprocessing outcomes, but we propose that elaboration is the most likely pre-cursor. This follows from Rafaeli’s (1988) conceptualization of interactivity.Under his definition, interactivity is “the extent to which communicationreflects back on itself, feeds on and responds to the past” (Newhagen andRafaeli 1996, 6). This can happen only when an individual relates incominginformation to other information and responds to it accordingly. Cognitiveelaboration must occur for a user to make a meaningful response to a messagereceived. Only in this case can the interactive behavior (i.e., message send-ing) of the user be characterized as truly interactive. The bold portions of Fig-ure 2 reflect this hypothesis:

Hypothesis 3: Greater interactive behavior should lead to a greater proportion ofcognitive elaboration.

The think-aloud method we will use to measure cognitive processing allowsus to determine exactly when cognitive elaboration occurs relative to interac-tive behavior. Therefore, we can test this hypothesis:

Hypothesis 4: Interactive behavior and cognitive elaboration should be coupledtemporally.

A previous study (Eveland and Dunwoody 2000) suggested that high lev-els of disorientation might be coupled with user inexperience with the Web,leading us to predict the following:

Hypothesis 5: Higher levels of Web use will be associated with a reduced need fororientation.

Finally, because elaboration has been demonstrated to improve storage of in-formation in memory, it follows that a higher level of truly interactive behav-ior (message sending coupled with cognitive elaboration) should do thesame. Therefore, we hypothesize the following:

Hypothesis 6: Users involved in greater amounts of interactive behavior shouldhave greater recall of Web site content.

Method

A think-aloud protocol was selected as a means of glimpsing informationprocessing strategies of individuals using the Web. Under this methodology,


users are instructed to express verbally the thoughts they are having as theynavigate through a Web site. This method of data collection has beendefended most vigorously by Ericsson and Simon (1993) and is used regu-larly in such fields as education (Calvi 1997; Hill and Hannafin 1997) andengineering (Carmel, Crawford, and Chen 1992; Darken and Sibert 1996).

Research on think-aloud protocols shows that the act of verbalizing doesnot greatly alter the actions or thoughts of the experimental participant(Ericsson and Simon 1993). As long as the instructions given to users arenondirectional, such as “try to think aloud, as if you were alone and talking toyourself,” the person’s thought processes should be relatively unaffected bytalking aloud. During the think-aloud process, participants were not asked toexplain their behaviors, as this has been found to change subsequentbehaviors.

The think-aloud method provides a verbal record of cognitive activity thatcan subsequently be coded for maintenance, elaboration, and orientationbehaviors.

Participants

The think-aloud procedure produces a large volume of verbal data andcoding per respondent, and analysis is lengthy. Because of this, a relativelysmall number of participants—twenty—was selected. Since statistical con-trols for individual differences that might affect user activity would be impos-sible for a sample of this size, we controlled for differences using a purposive,rather than random, sample. Participants were drawn from those whoresponded to advertisements in a medium-sized Midwestern city. They werecompensated for their time. Equal numbers of male and female participantswere selected.

To control for participant interest, one topic, science, was selected as thecontent area for study. Participants were asked to rate their interest in varioussubjects on a 10-point scale. Those who responded with 7 or above on sciencewere placed in a candidate pool from which the final twenty study partici-pants were selected based on the criteria outlined above. The twenty partici-pants rated their interest in science a mean of 8.7, compared to a mean of 6.1for all callers.

Respondents reported an average of 11.9 hours of Web use in a typicalweek. We selected participants with high Web use (defined here as 15 or morehours per week) and low use (7 or fewer hours, but not zero). Our resultinglow-use group had a 2.7-hour mean and our high-use group a 25.0-hourmean.


The participants had a minimum of “some college,” most had B.A.degrees, and a few had earned graduate degrees. Ten participants were in the$20,000 to $40,000 income bracket, nine in the less than $20,000 bracket,and one in the $60,000 to $80,000 bracket. None of the participants were cur-rently undergraduates; four were in graduate or professional degree pro-grams, and the rest were employed. Participants ranged in age from earlytwenties to late forties. The mean age range for each study condition wasthirty-five to thirty-nine.

Conditions

Study participants were randomly assigned (by last digit of home phonenumber) to either the high or the low “user options” Web condition. Thelow-option Web site selected for study was the Why Files (www.whyfiles.org), a Web magazine devoted to science topics currently in the news. At thetime of data collection, stories on the Why Files were primarily presented astext, with a small number (typically five or less) of link options per screen.Rarely on this site was a user asked to use his or her keyboard or mouse to pro-vide information that would alter screen content.

The Exploratorium Museum of Science, Art and Human Perception Website (www.exploratorium.edu) was selected as the high-option condition.Web visitors are often asked to use their keyboard and mouse to affecton-screen changes in content. Stories on the Exploratorium site contain alarge number of hypermedia links per screen (typically a dozen or more).Many stories use interactive modules that allow readers to key in data that areincorporated in the presentation.

The content of these sites is similar but certainly not equivalent. Each site,for example, had a story about the solar max phenomenon. There werenumerous other areas of content overlap as well.

Procedures

The twenty selected subjects participated individually in a thirty-minuteWeb surfing session during which they were asked to think aloud. Each indi-vidual first practiced the think-aloud technique on a number of warm-upexercises, including a visit to the Web site www.howstuffworks.com. Eachparticipant used either a PC or an iMac computer and either Netscape or theInternet Explorer Web browser. The computers were connected to theInternet by a high-speed Ethernet line. A VCR was used to record the screenimages during the Web browsing exercise, and the voice of the participant


was captured simultaneously. A clock on the computer screen displayed thetime in seconds so that instances of user activity and resulting cognitivebehavior could be temporally isolated.

Study participants started at the home page of their assigned Web site.They were allowed to follow any links, including those that took them toother Web sites. Most participants who did this returned without guidance,but one became lost and requested assistance in returning to the original Website. The principal investigator supplied that user with the appropriate Webaddress, and the participant continued the session without further difficulty.

At the conclusion of the session, each participant was asked about onestory or area of the Web site where he or she had spent at least 3 minutes(mean of 7.5 minutes) during the middle of the session. Study participantswere never asked about the last 5 minutes of their sessions, as recall should behigher with a short interval between exposure and questioning. For this por-tion of the study, results from four of the twenty participants could not beused because the users either never spent more than 3 minutes on one sectionor spent the entire session on one part of the assigned Web site. The total num-ber of words spoken during recall and the total number of words about sitecontent (as opposed to navigation) were used as recall measures.

Operational Definitions: Information Processing

The key variable of interest—information processing style—was deter-mined based on the comments made by each participant during thethink-aloud session. For each ten-second interval, the user’s comments werecoded as rehearsal, elaboration, or orientation.

The rehearsal code was selected when a study participant merely readfrom the screen or rephrased the information without the addition of otherfacts or personal opinions.

Elaboration was defined as any comment demonstrating a connection ofcurrently encountered information to prior knowledge, including connec-tions made to other information from this episode of Web use. This alsoincludes any affective evaluation by the user, such as “that’s not true,” or “Ilike this topic,” as such statements assume prior knowledge or experience bythe study participant.

Orientation was coded whenever a user made a comment about site navi-gation, such as “let’s try this and see where it takes us,” “I’m lost,” or “nowI’ve figured out where I am.”


Coding and Reliability

Every ten-second interval of each participant’s thirty-minute Web brows-ing session was coded for Web site (Why Files, Exploratorium, or other), useractivity (keyboard inputting, mouse clicking, other mouse action not includ-ing scrolling, or none), information processing (rehearsal, elaboration, or ori-entation), and domain (a comment about site content vs. one about site struc-ture or navigation). If more than one category for a variable occurred duringthe ten-second interval, the one that occurred for a majority (51 percent) ofthe interval was considered the proper code. A “mixed” code was used foreach variable when multiple behaviors prevented any category from reaching51 percent.

A reliability test involving the principal investigator and another coder (apaid doctoral student) was conducted on the coding scheme. As the variablesare all nominal, Scott’s π was used to calculate intercoder reliability. Reli-ability for the coding scheme was .83 overall, with individual variables of .75for information processing, .78 for domain, .85 for user activity, and .93 forWeb site.

Results

Action Phase

The two Web sites were selected to maximize differences in interactivebehavior while minimizing differences in content. The Exploratorium site,with content more elaborately hyperlinked and with more opportunities foruser input, was selected on the assumption it would lead to more mouse andkeyboard activity than the Why Files site. We also selected research partici-pants with high and low Web use on the assumption that their interactivebehavior levels would vary accordingly. If these premises were correct, wewould expect the interactive behavior levels corresponding to those inTable 1.

The results validate these assumptions. Table 2 shows the percentage oftime intervals during which users were engaged in activity that changedscreen content.

The results demonstrate a sizable difference in interactive behavior of par-ticipants who spent time on the Why Files site (26 percent overall) comparedto those who used the Exploratorium site (47 percent). This difference is


significant (χ2 = 93.0, df = 1, p < .001) and confirms hypothesis 1. The moremodest difference in interactive behavior for those with high (40 percent) andlow (34 percent) Web use is also statistically significant (χ2 = 8.4, df = 1, p <.01) and confirms hypothesis 2. The action phase of our proposed model issupported by the data.

Cognition Phase

Hypothesis 3 predicts that higher levels of interactive behavior amongusers should encourage cognitive elaboration during Web use. Table 3 showsthe percentage of comments coded as elaborations for each user group.

These data exclude time intervals when a participant followed a link to anoutside page and those time intervals that could not be coded as rehearsal,elaboration, or orientation. The proportions are similar to the interactivebehavior levels shown in Table 2 with the exception of the low-use Why Filesgroup. If our hypothesis had been correct, this cell would contain the smallest


TABLE 1Predicted Interactive Behavior Levels

Exploratorium Why Files

High Web use Highest MiddleLow Web use Middle Lowest

TABLE 2Interactive Behavior Levels (in percentages)


High Web use 51 28Low Web use 43 24

TABLE 3Elaboration Levels (in percentages)


High Web use 54 35Low Web use 47 48

elaboration value instead of the second largest. This may be due in part to oneparticipant in that cell whose behavior was very unusual compared to allother study participants (the participant’s comments were coded as 82 per-cent elaborative). Excluding this case from the data as a statistical flier wouldreduce that subgroup to 40 percent and yield a positive correlation betweeninteractivity and elaboration; however, that correlation would be very modest(Pearson’s r = .11, n = 20).

A post hoc analysis of the effect of Web site structure and Web experienceon cognitive processing is illuminating. Slightly more than half (51 percent)of the comments made by Exploratorium Web users could be categorized aselaborations compared to 42.5 percent for the Why Files users. The differ-ences in information processing strategies between the two groups were sta-tistically significant (see Table 4).

Users of the Exploratorium site spent slightly more time in orientationbehavior than did those using the Why Files site, a pattern consonant with theExploratorium’s more complex structure, while those on the Why Files spentmore time on rehearsal than did those using the Exploratorium site.

For Web experience, contrary to expectations, high-use study participantsactually engaged in less cognitive elaboration than did the low-use group (seeTable 5).

We offer possible explanations for these contradictory findings in the con-clusions that follow.

While an overall increase in elaboration was recorded for users of themore interactive Web site, this cognitive work did not occur more often dur-ing interactive behavior, as predicted in hypothesis 4. An examination of theten-second time intervals during which interactive behavior occurred revealsa large increase in orientation at those times, perhaps not surprising since thisusually involved the traversing of hyperlinks (see Table 6).

The increased orientation that occurs when users are navigating a site doesreduce the other forms of information processing examined here (a cognitiveload effect), but elaboration decreases only slightly while rehearsal is almosthalved. There is almost a one-to-one correspondence between increased ori-entation and decreased rehearsal during time intervals involving interactivebehavior. This result is the same when we look at the two Web site groupsseparately (see Table 7).

Again we see large increases in orientation during user activity coupledwith large decreases in rehearsal and smaller decreases in elaboration. Thereis no support for hypothesis 4, but a competing hypothesis, that interactivebehavior and cognitive orientation are temporally connected, is suggested bythe data.


Hypothesis 5 predicts that more experienced Web users will experience adecreased need for cognitive orientation. An analysis of all time intervals (N =2,279) reveals almost no difference between high- and low-use groups (seeTable 5). However, further analysis of the data does reveal an interesting dis-tinction between the high- and low-Web-use groups concerning cognitiveorientation. While the groups are virtually identical when not engaged in key-board or mouse activity (orientation accounts for 13 percent of the behaviors


TABLE 4Information Processing by Web Site (in percentages)

Exploratorium Why Files(n = 1,016) (n = 1,263)

Rehearsal 26 39Elaboration 51 42.5Orientation 23 18.5Total 100 100

NOTE: χ2 = 43.7, df = 2, p < .001.

TABLE 5Information Processing by Web Use (in percentages)

High Use Low Use(n = 1,010) (n = 1,269)

Rehearsal 38 29Elaboration 42 50Orientation 20 21Total 100 100

NOTE: χ2 = 20.5, df = 2, p < .001.

TABLE 6Information Processing by Interactive Behavior (in percentages)

Activity Nonactivity(n = 992) (n = 1,741)


NOTE: χ2 = 213.5, df = 2, p < .001.

of both groups), there is a difference in their orientation behavior during peri-ods of interactive behavior (see Table 8).

These results, while suggestive of a difference between the high- andlow-use groups, are not statistically significant at the .05 level.

The final hypothesis concerns the recall of information by study partici-pants. Users spent varying amounts of time on the sections of the Web sitethey were subsequently asked to recall. Since we would expect some correla-tion between time spent and the amount of information that was laterrecalled, we have adjusted the data here by dividing them by the number ofminutes a user spent on the part of the Web site in question.

As with their Web sessions as a whole, the Exploratorium users hadgreater interactive behavior levels (2.8 actions per minute) on the targetedrecall sections compared to the Why Files group (1.4 actions per minute).This result is statistically significant (t = 3.29, p < .01).

The amount of cognitive elaboration on these sections is also higher forthe Exploratorium group (3.3 elaboration intervals per minute) compared tothe Why Files group (1.9 intervals) at a statistically significant level (t = 2.91,p < .01).


TABLE 7Information Processing by Web Site by Interactive Behavior


Activity Nonactivity Activity Nonactivity

Rehearsal 21 30 21 45Elaboration 48 53 36 45Orientation 31 17 42 10

TABLE 8Information Processing during Interactive

Behavior by Web Use (in percentages)

High Use Low Use(n = 384) (n = 409)


NOTE: χ2 = 5.0, df = 2, p = .08.

As predicted, the total amount recalled is also higher for theExploratorium group (45.1 words per minute of attention) compared to theWhy Files participants (26.2 words) at a significant level (t = 2.04, p < .05). Asimilar difference is found when we look only at content-specific recall (33.6vs. 18.7 words per minute of attention).

Additional Findings

Our data suggest a substantial decrease in cognitive orientation from asimilar study conducted three years earlier (Eveland and Dunwoody 2000).Our findings indicate a substantial decrease, from nearly 40 percent in 1997to 22 percent in 2000 (see Table 9).

The results are statistically significant. Although the two studies use dif-ferent units of analysis (individual thoughts vs. ten-second intervals), we donot believe the large differences in the proportions of each information pro-cessing category can be explained by that fact.

Conclusions

The data presented here support the action phase of the interactiveinformation-processing model we proposed but require a reconsideration ofthe cognition phase. Users of the more complex site engaged in more interac-tive behavior and demonstrated greater levels of cognitive elaboration andsubsequent recall of content. But a correlation between interactive behaviorand elaboration across all twenty participants was not found, nor was a tem-poral connection found between these variables. There are a number of possi-ble explanations, the first of which goes to study design.

In this quasi-experimental study, we chose to use actual Web sites ratherthan simulations and allowed participants great freedom in exploring them.


TABLE 9Information Processing by Year (in percentages)

1997 (n = 2,790) 2000 (n = 2,733)

Rehearsal 23 30Elaboration (including evaluations) 38 48Orientation 39 22Total 100 100

NOTE: χ2 = 191, df = 2, p < .001.

This type of experience more closely mirrors actual media use and affords usgreater external validity. But the cost is internal validity. The two sites, whilesimilar in content, are certainly not the same. Variables other than interactiv-ity may be responsible for increases in elaboration and recall found for theusers of the Exploratorium site.

Higher levels of Web use were found not to promote cognitive elaborationbut perhaps to hinder it. This was especially apparent for high-use partici-pants assigned to the less complex Web site (Why Files). This group recordedthe lowest elaboration levels. It may be that like television channel surfers,high-use Web browsers have a lower involvement level with relatively simplecontent. High-use participants assigned to the more complex site(Exploratorium) recorded the highest elaboration levels.

It is possible that involvement or attention is an intervening variable thatneeds to be accounted for in our model. And, it may be the case that what wewould term true interactivity (interactive behavior coupled with elaboration)and meaningless interactivity (surfing or browsing behavior not coupled withelaboration) are both occurring. In this case, both viewpoints on the value ofinteractivity, discussed in the introduction to this article, could have merit.

As predicted in previous research in this area (Eveland and Dunwoody2000), orientation behavior among research participants has declined overtime. In both studies, research participants were average citizens, not collegeundergraduates. As the general population becomes more accustomed tousing the Web, we would expect this trend to continue.

Finally, and most promising for proponents of Web-based informationsystems, this exploratory study did find evidence of a greater amount ofcontent-specific recall for participants assigned to the more interactive Website. This result warrants further investigation. A more controlled experiment—one in which the content does not vary from participant to participant—would be a logical next step.

References

Anderson, J. 1990. Cognitive psychology and its implications. San Francisco: Freeman.Anderson, J., and L. Reder. 1979. An elaborative processing explanation of depth of processing.

In Levels of processing in human memory, edited by L. Cermak and F. Craik, 385-403.Hillsdale, NJ: Lawrence Erlbaum.

Baker, L. 1989. Metacognition, comprehension monitoring and the adult reader. EducationalPsychology Review 1:3-38.

Ball-Rokeach, S. J., and K. Reardon. 1988. Monologue, dialogue, and telelog. In Advancingcommunication science: Merging mass and interpersonal processes, edited by R. P. Hawkins,J. M. Wiemann, and S. Pingree, 135-61. Newbury Park, CA: Sage.


Bandura, A. 1994. Social cognitive theory of mass communication. In Media effects: Advancesin theory and research, edited by J. Bryant and D. Zillmann, 61-90. Hillsdale, NJ: LawrenceErlbaum.

Belmont, J. M., and E. C. Butterfield. 1971. Learning strategies as determinants of memory defi-ciencies. Cognitive Psychology 2:411-20.

Berry, C. 1983. Learning from television news: A critique of the research. Journal of Broadcast-ing 2:359-70.

Booth, A. 1970-1971. The recall of news items. Public Opinion Quarterly 34:604-10.Bordewijk, J. L., and B. van Kaam. 1986. Towards a new classification of tele-information ser-

vices. Intermedia 14 (1): 16-21.Burbules, N., and T. Callister. 1996. Knowledge at the crossroads: Some alternative futures of

hypertext learning environments. Educational Theory 46 (1): 23-51.Calvi, L. 1997. Navigation and disorientation: A case study. Journal of Educational Multimedia

and Hypermedia 6:305-20.Carmel, E., S. Crawford, and H. Chen. 1992. Browsing in hypertext: A cognitive study. IEEE

Transactions on Systems, Man, and Cybernetics 22:865-84.Darken, R. P., and J. L. Sibert. 1996. Navigating large virtual spaces. International Journal of

Human-Computer Interaction 8 (1): 49-71.DeFleur, M. L., L. Davenport, M. Cronin, and M. DeFleur. 1992. Audience recall of news stories

presented by newspaper, computer, television and radio. Journalism Quarterly 69:1010-22.Ericsson, K. A., and H. A. Simon. 1993. Protocol analysis: Verbal reports as data. Cambridge,

MA: MIT Press.Estes, W. K. 1988. Human learning and memory. In Stevens’handbook of experimental psychol-

ogy, edited by R. C. Atkinson, R. J. Herrnstein, G. Lindzey, and R. D. Luce, chap. 5. NewYork: John Wiley.

Eveland, W. P., and S. Dunwoody. 2000. Examining information processing on the World WideWeb using think aloud protocols. Media Psychology 2:219-44.

Foltz, P. 1996. Comprehension, coherence, and strategies in hypertext and linear text. In Hyper-text and cognition, edited by J. -F. Rouet, J. Levonen, A. Dillon, and R. J. Spiro, 109-36.Mahwah, NJ: Lawrence Erlbaum.

Fredin, E. 1997. Rethinking the news story for the Internet: Hyperstory prototypes and a modelof the user. Journalism and Mass Communication Monographs 163.

Furnham, A., and B. Gunter. 1989. The primacy of print: Immediate cued recall of news as afunction of the channel of communication. Journal of General Psychology 116:305-10.

Gardner, H. 1985. The mind’s new science. New York: Basic Books.Gibson, J. J. 1966. The senses considered as perceptual systems. Boston: Houghton Mifflin.Goffman, E. 1967. Interaction ritual: Essays on face-to-face behavior. Chicago: Aldine.Greene, R. L. 1992. Human memory: Paradigms and paradoxes. Hillsdale, NJ: Lawrence

Erlbaum.Ha, L., and E. L. James. 1998. Interactivity reexamined: A baseline analysis of early business

Web sites. Journal of Broadcasting and Electronic Media 42:457-74.Halpern, D. H. 1997. The impact of different operationalizations of the media on political knowl-

edge and learning: A meta-analytic review. Paper presented at the annual meeting of theInternational Communication Association, Montreal, Canada, May.

Hamilton, R. J. 1997. Effects of three types of elaboration on learning concepts from text. Con-temporary Educational Psychology 22:299-318.

Hawkins, R. P., and S. Pingree. 1997. Measuring time frames of communication behaviors incomputer use. Paper presented at the annual meeting of the International CommunicationAssociation, Montreal, Canada, May.


Hill, J. R., and M. J. Hannafin. 1997. Cognitive strategies and learning from the World WideWeb. Educational Technology Research and Development 45 (4): 37-64.

Kipper, P. 1991. The computer television marriage. Television Quarterly 25 (3): 13-19.Landow, G. P. 1991. Position statement for Panel on Structure, Navigation and Hypertext: The

status of the navigation problem. In Hypertext ’91: Proceedings of the third ACM conferenceon hypertext, 364. New York: Association for Computing Machinery.

Lawless, K., and J. M. Kulikowich. 1996. Understanding hypertext navigation through clusteranalysis. Journal of Educational Computing Research 14 (4): 385-99.

Marchionini, G. 1988. Hypermedia and learning: Freedom and chaos. Educational Technology28 (11): 8-12.

Mayes, T., M. Kibby, and T. Anderson. 1990. Learning about learning from hypertext. InDesigning hypermedia for learning, edited by D. H. Jonassen and H. Mandl, 227-50. NewYork: Springer-Verlag.

McDaniel, M., and G. Einstein. 1989. Material appropriate processing: A contextualist approachto reading and study strategies. Educational Psychology Review 1:113-46.

McDonald, S., and R. J. Stevenson. 1996. Disorientation in hypertext: The effects of three textstructures on navigation performance. Applied Ergonomics 27 (1): 61-68.

McMillan, S. 1999. Health communication and the Internet: Relation between interactive char-acteristics of the medium and site creators, content and purpose. Health Communication 11(4): 375-90.

Neuman, W. R., M. R. Just, and A. N. Crigler. 1992. Common knowledge: News and the con-struction of political meaning, 78-109. Chicago: University of Chicago Press.

Newhagen, J. 1997. The role of feedback in the assessment of news. Information Processing andManagement 33 (5): 583-94.

Newhagen, J., J. Cordes, and M. Levy. 1995. [email protected]: Audience scope and the per-ception of interactivity in viewer mail on the Internet. Journal of Communication 45 (3):164-75.

Newhagen, J., and S. Rafaeli. 1996. Why communication researchers should study the Internet:A dialogue. Journal of Communication 46:4-13.

Ogan, C. 1993. Listserver communication during the Gulf War: What kind of medium is theelectronic bulletin board? Journal of Broadcasting and Electronic Media 3:177-96.

Price, V., and J. Zaller. 1993. Who gets the news? Alternative measures of news reception andtheir implications for research. Public Opinion Quarterly 57:133-64.

Rafaeli, S. 1988. Interactivity: From new media to communication. In Advancing communica-tion science: Merging mass and interpersonal processes, edited by R. P. Hawkins, J. M.Wiemann, and S. Pingree, 110-34. Newbury Park, CA: Sage.

Rafaeli, S., and F. Sudweeks. 1997. Networked interactivity. Journal of Computer-MediatedCommunication 2 (4). Available: http://www.ascusc.org/jcmc/vol2/issue4/rafaeli.sudweeks.html.

Rice, R. 1988. Issues and concepts in research on computer-mediated communication systems.Communication Yearbook 12:436-76.

Rice, R. E., and F. Williams. 1984. Theories old and new: The study of new media. In The newmedia: Communication, research, and technology, edited by R. E. Rice, 55-80. BeverlyHills, CA: Sage.

Rouet, J. -F., and J. Levonen. 1996. Studying and learning with hypertext: Empirical studies andtheir implications. In Hypertext and cognition, edited by J. -F. Rouet, J. Levonen, A. Dillon,and R. J. Spiro, 9-24. Mahwah, NJ: Lawrence Erlbaum.


Rumelhart, D. E., P. H. Lindsay, and D. A. Norman. 1972. A process model for long-term mem-ory. In Organization of memory, edited by E. Tulving and W. Donaldson. New York: Aca-demic Press.

Salomon, G. 1990. Cognitive effects with and of computer technology. CommunicationResearch 17 (1): 26-44.

Shaw, R. E., and J. Bransford. 1977. Perceiving, acting and knowing. Hillsdale, NJ: LawrenceErlbaum.

Shoemaker, P., C. Schooler, and W. A. Danielson. 1989. Involvement with the media: Recall ver-sus recognition of election information. Communication Research 16:78-103.

Simon, H. E., and L. Siklossy. 1972. Representations and meaning: Experiments with informa-tion processing systems. Englewood Cliffs, NJ: Prentice Hall.

Steuer, J. 1992. Defining virtual reality: Dimensions determining telepresence. Journal of Com-munication 42:73-93.

Sundar, S. S., J. Brown, and S. Kalyanaraman. 1999. Reactivity vs. interactivity: Impression for-mation effects of message contingency in political Web sites. Paper presented at the annualmeeting of the International Communication Association, San Francisco, May.

Tripp, S. D., and W. Roby. 1990. Orientation and disorientation in a hypertext lexicon. Journal ofComputer-Based Instruction 17 (4): 120-24.

Walther, J. B. 1994. Anticipated ongoing interaction versus channel effects on relational com-munication in computer-mediated interaction. Human Communication Research 20:473-501.

Westley, B. H., and M. MacLean. 1957. A conceptual model for communication research. Jour-nalism Quarterly 34:31-38.

Wicks, R. H. 1992. Schema theory and measurement in mass communication research: Theoret-ical and methodological issues in news information processing. In Communication Year-book 15, edited by S. A. Deetz, 115-45. Newbury Park, CA: Sage.

Williams, F., R. E. Rice, and E. M. Rogers. 1988. Research methods and the new media. NewYork: Free Press.

Woodall, W. G., D. K. Davis, and H. Sahin. 1983. From the boob tube to the black box: Televi-sion news comprehension from an information processing perspective. Journal of Broad-casting 27:1-23.

MARK TREMAYNE is a doctoral student in the School of Journalism and Mass Commu-nication at the University of Wisconsin–Madison. His research focuses on the role ofinteractivity and nonlinearity in Web-based communication.

SHARON DUNWOODY is the Evjue-Bascom Professor and the director of the School ofJournalism and Mass Communication at the University of Wisconsin–Madison. She con-ducts research on various aspects of communicating science to the public and teachesscience writing, among other topics.


SCIENCE COMMUNICATIONLogan / SCIENCE MASS COMMUNICATIONThis article provides a conceptual history of science mass communication, which is seen asdivided into the scientific literacy and interactive science traditions. The origins of the ideas thatunderlie the scientific literacy and interactive science traditions, as well as some of the issuesresearchers have raised, are introduced. The author argues the two traditions are not mutuallyexclusive, although the interactive tradition is a response to the applied problems within the sci-entific literacy model. It is argued that the pace of research might be accelerated if there were amore comprehensive collaboration among science communication, health communication, andrisk communication scholarship.

Science Mass CommunicationIts Conceptual History

ROBERT A. LOGANUniversity of Missouri–Columbia

In a recent essay, Einsiedel and Thorne (1999) found the underpinnings ofprevious research about the public understanding of science is divided intotwo broad areas: (1) the public’s scientific literacy and (2) an interactive sci-ence model.

Einsiedel and Thorne (1999) explained that studies representing the scien-tific literacy model “suggest that knowledge of particular basic scientificideas and concepts is required for people to function well in a variety of cul-tural contexts. Scientific knowledge within this framework is generally por-trayed as fixed and certain” (p. 49). In terms of mass communication, the sci-entific literacy model is implied to be pedagogically based, a flow ofknowledge from the scientific community through the press to citizens. Incontrast, the interactive science model “takes as a given the uncertaintiesembedded in the scientific enterprise and the idea that science cannot be sepa-rated from its social and institutional connections” (Einsiedel and Thorne

Author’s Note: Address correspondence to Robert A. Logan, School of Journalism, Universityof Missouri, 120 Neff Hall, Columbia, MO 65211; phone: 573-882-4714; fax: 573-884-5400;e-mail: [email protected] or [email protected].


135

1999, 50). Besides acknowledging uncertainty and a science-society connec-tion, the interactive science model suggests the flow of science knowledge isnot always from experts to laypersons and implies it might be more shared ormultidirectional.

Both models, or underlying conceptions, of how science is communicatedto the public are seen as rough “portraits” of the process by which science’smass communication conceptually evolved (Einsiedel and Thorne 1999).

As Einsiedel and Thorne (1999) suggested, science communication’sresearch models and traditions provide insights into the history of the fieldand its future direction. By reviewing them, the foundations of the field andits sense of direction are clarified, and the field of science communication’slinks to related fields (such as health and risk communication) and the historyof mass communication research become more apparent. While Einsiedeland Thorne’s discussion continued in terms of how to communicate uncer-tainty to different publics or audiences, their scientific literacy and interactivemodels are used more broadly here to compare how research about science’smass communication is undergirded, the enthusiasms of contemporaryresearchers, and their topical range and future challenges. Since a compre-hensive literature review of research findings is included elsewhere in thisissue of Science Communication (Weigold 2001 [this issue]), this article pro-vides a conceptual history of science mass communication, introduces theideas that developed each model, and discusses some of the subsequentissues researchers raised.

The scientific literacy and interactive models are seen as a foundation toexplore research traditions because each begins with a similar challengeregarding improving the public understanding of science, while each raisesseparate, initial conceptual assumptions that foster different research direc-tions and insights. To pose a comparison, the scientific literacy tradition firstwill be introduced with an emphasis on its initial assumptions and the rangeof research topics about audiences, sources, messages, and channels thatemerged from this model. Since the historical origins of the scientific literacytradition are rarely discussed in the literature, some attention is paid to itsdevelopment during the first thirty years of the twentieth century. The con-ceptual roots of the transition to the interactive model in the 1980s are dis-cussed, as are the presumptions embedded in the comparatively new andevolving interactive science tradition.

This article argues that the two traditions are not mutually exclusive,although the interactive tradition is a response to the applied problems withinthe science literacy model. It is argued that the pace of research might beaccelerated if there were a more comprehensive collaboration among sciencecommunication, health communication, and risk communication scholarship.


The terms research “model” and research “tradition” are used inter-changeably, even though a “model” sometimes is interpreted as a diagram ofresearch constructs while a research “tradition” sometimes refers to a muchbroader conceptual direction. Unlike scientific areas that Rowan (1999)described as textbook science, the subdiscipline of science mass communica-tion and the broader field of mass communication research represent emerg-ing, pioneer, and somewhat unsettled social sciences (McQuail 2000;Severin and Tankard 1997). As a result, both the narrowed topic of science’smass communication and its larger disciplinary context provide rough mod-els partially supported by preliminary qualitative and quantitative evidencethat, in aggregate, converge into traditions or broad research directions. Theideas within this article, I hope, explore the emerging and dynamic spirit ofeach tradition and introduce how science communication research has beenconceived, is evolving, and might progress.

Scientific Literacy Tradition: The ClassicalModel and Audience-Based Research

Unlike many areas within mass communication, the public communica-tion of science originally was grounded in a pedagogical purpose. This sec-tion describes how the classical model of science communication evolvedhistorically and how research within the scientific literacy tradition encom-passed public understanding and the publics that converge on science. Thescientific literacy tradition also encompasses areas of inquiry regarding thesources of science news, how science gets transformed from research resultsinto news stories, and the impact of subtle news editing and writing decisionson science policy, public affairs, and public opinion. These topics are dis-cussed in a separate section below.

Tobey (1971) described the almost-missionary zeal of a few leading scien-tists during the first thirty years of the twentieth century, who sought toimprove the capacity of Americans to make rational public affairs decisionsabout science and to better integrate scientific knowledge to improve thequality of their lives. Tobey explained that Slossen, Heyl, Millikan, and Halewere dedicated to a complex agenda that included (1) cultivating the idea oflifelong learning for citizens, (2) helping persons live healthier and longerlives by promoting scientific awareness, (3) encouraging support for the sci-entific method as a strategy for public officials to assess complex publicaffairs choices, (4) helping citizens and public officials better understand theconnection between investment in science research and the United States’economic future, (5) improving public investment in science, (6) fostering

Logan / SCIENCE MASS COMMUNICATION 137

more interest in science as a career among American youths, (7) enhancingpublic goodwill and support for science among taxpayers, and (8) nurturing apublic will to support science as a nonpartisan staple of national investmentin the future of America’s economy and culture.

Slossen, Heyl, Millikan, and Hale also believed understanding sciencerepresented an inherently transformative, highly personal experience (Tobey1971). A commitment to the public understanding of science not only pro-vided immediate rewards of improving quality of life and elevating publicaffairs deliberations but also was perceived as integral to the elevation ofdemocracy, culture, and the evolution of human potential (Tobey 1971). Sim-ilar to Bronowski (1973) or Silver (1998), Slossen, Heyl, Millikan, and Haledid not accept arguments that science is deterministic or is dehumanizing tohumanistic traditions in the arts and culture, as Snow (1993) delineated a gen-eration later. Besides longer lives, improved public health, and understandingof nature, Slossen, Heyl, Millikan, and Hale, whom Tobey (1971) termed“national scientists,” believed the growth of science was tied to whatBronowski later termed “the ascent of man.” Bronowski argued that the evo-lution of science (more than most other historical developments) fostered theadvance of democratic national institutions, the growth of major universities,the importance of public education, the growth of the middle class around theworld, and an increased confidence for individuals to challenge conventionalwisdom and dogma, and that it engendered a humane passion to probe fortruth and evidence. To Slossen, Heyl, Millikan, and Hale, science’ssociocultural and individual influences surpassed its immediate findings andapplications (Tobey 1971). Science’s influence, as Bronowski (1973), Silver(1998), and Holton (1993) later maintained, elevated human confidence andcapacity to create a modern, industrial, self-critical, tolerant, and democraticsociety.

From the outset, Slossen, Heyl, Millikan, and Hale also were concernedabout the poor levels of public education about science in the United States(Tobey 1971). They believed the public’s poor educational foundation wasexacerbated by an entertainment-oriented popular culture advanced by thenews and mass media (Tobey 1971). While science fiction and reinforcingpopular superstitions remained ubiquitous (e.g., the depiction of mad andeccentric scientists in popular fiction), the news and mass media were seen asproviding minimal exposure to understanding actual scientific findings(Tobey 1971). Tobey (1971) explained how Slossen, Heyl, Millikan, andHale orchestrated reform by working with journalists to publish news storiesabout breaking science advances and to provide occasional explanations ofscience and medical processes, such as how a heart functions or why citizensshould care about the then-new Einstein theory of relativity. Some of the


national scientists during the first thirty years of the twentieth century urgednews magnates, especially E. W. Scripps, to institutionalize science reportingas an ongoing news beat. Slossen, Heyl, Millikan, and Hale approachednewspaper publishers with a vision about how their coverage of sciencecould safeguard the nation’s future and generate more credibility for the press(Tobey 1971).

It should be noted that Slossen, Heyl, Millikan, and Hale’s ideas wereoften expressed more in personal letters, speeches, and popular editorialsthan in formal scholarship. However, Slossen and Downey (1922) proposedan empirically based methodology to demonstrate that the transfer of scien-tific ideas could increase a person’s learning curves and inspire “creativeimagination.”

Although Slossen, Heyl, Millikan, and Hale’s conceptual framework wasoutlined more than seventy years ago, it represents a classic metaphor for theprocess and effects of science’s mass communication that infused scholar-ship throughout the twentieth century. Some contemporary publications,notably a book by Hartz and Chappell (1997), propose a parallel view of sci-ence’s mass communication process and update the social and economicimpact of poor public understanding of science plus the relative roles andresponsibilities that scientists and journalists share to elevate public under-standing. Hartz and Chappell advanced an informal framework similar to thenational scientists’, wherein a diverse, socially therapeutic impact of sci-ence’s mass communication is subsumed. Consequently, scientists and jour-nalists are asked to provide leadership to improve both the availability of sci-ence news to citizens and the quality of the information the public receives(Hartz and Chappell 1997).

A portion of science communication’s classical framework was indirectlyadopted by health communication campaign research in the 1970s.Bandura’s (1977) social learning model conceived the news and mass mediaas conveyors of targeted health messages from health care providers throughcommunicators to patients that increased public knowledge about specifiedhealth habits. Increased public information was linked to improvements intherapeutic health care behaviors, which fostered more popular appreciationand goodwill for the sources of the messages. A series of studies at StanfordUniversity in the 1970s compared lifestyle habits and cardiovascular knowl-edge between control and target audiences, and it empirically assessed a por-tion of the linear classical science communication flow model informallyadvanced fifty years before (Farquhar, Magnus, and Maccoby 1981). Morerecently, Wallack (1993) took aspects of the social learning model as a primerfor public interest health organizations to generate news media attention andobtain publicity for their health messages. While health communication


literature based on social learning models does not credit or cite the classicscience communication model, there are conceptual consistencies betweenthe two disciplines.

Although contemporary researchers adopted alternative models of sci-ence’s mass communication and adapted the classical science communica-tion model to other purposes, it is noteworthy that for most of the twentiethcentury, the linear linkages and assumptions embedded within the classicalframework were decades ahead of conceptual developments within masscommunication research. The early, comprehensive models that projectedthe impact of sources (scientists), messages (the news media’s content), andmedia channels (such as newspapers, magazines, radio) on readers were notadvanced until the late 1940s—more than twenty-five years after Slossen,Heyl, Millikan, and Hale discussed their informal, conceptual framework.Applied to the flow of news, Shannon and Weaver (1949) described a linearprocess where information (the intent and original content) of news sourcesencountered entropy (or loss of intended meaning) within messages as theymoved from an original source through reporters, editors, news organiza-tions, and media channels to citizens. Westley and MacLean (1957) added theidea that feedback loops between sources, the content and context of newsmessages, and media channels were essential conceptual components withinthe same process. Much of the research in science communication during thepast thirty years has broadly explored Westley and MacLean’s conceptualadditions, which are discussed in the next section.

Besides sources, messages, and channels, Westley and MacLean (1957)emphasized that assessing the impact of news on “receivers” (the audiencefor news—viewers, listeners, and readers) was integral to the assessment ofmass communication’s social impact. Through the early 1960s, MacLean(1965, 1967) urged mass communication researchers to assess publiccognitions (what citizens know about public affairs topics and issues), atti-tudes (what citizens perceived about news, social institutions, and publicaffairs), and what is today termed a “conative dimension” (how personsintended to behave on the basis of their knowledge and attitudes). Bandura’s(1977) social learning model, which was advanced in the same era, provideda similar three-part division to assess audience responses to health communi-cation campaigns and similarly argued that broader mass communicationresearch should be audience centered. While Bandura and MacLean often areconsidered pioneers in advocating attention to the audience for news andmass media, Slossen, Heyl, Millikan, and Hale’s classical frameworkembraced the importance of understanding the audience for science aboutforty years earlier. From its outset, the classic metaphor of science communi-cation placed an emphasis on assessing how well basic scientific ideas and


concepts enabled the public to “function well in a variety of cultural con-texts,” as Einsiedel and Thorne (1999, 49) described.

In a series of essays about improving the research about the public’sunderstanding of science, Stephenson (1973) combined classical traditionswith then-contemporary ideas and suggested two audience-based approaches.Stephenson suggested researchers should investigate (1) what citizens knowabout science and (2) how persons perceive science. The first research direc-tion could assess the cognitive impact of science reporting after persons wereexposed to science news and other media content. The contrasting, secondresearch direction could anticipate how audiences might respond to futuremessages through a better understanding of audience attitudes about science,their attentiveness to public affairs, their attitudes about the news media, andother issues. The second research direction assumed a science reader, viewer,or listener is not a tabula rasa but represents a complex blend of prior knowl-edge, attitudes, and habits. The admixture of prior knowledge, attitudes, andhabits forms predispositions toward science and science news. Predisposi-tions affect how persons perceive the credibility of news sources; the extent towhich adults and children are interested in learning; possible motivations toread, listen, or watch science news; and potential recall of facts or conceptswithin a science story. By understanding predispositions and defining anycommon clusters of predispositions within audiences, Stephenson believedscience communicators could better tailor messages to suit different audi-ence needs.

Correspondingly, one contemporary branch of audience-based research inthe scientific literacy tradition explores cognitions by testing the public’s sci-ence knowledge. Scholars such as Miller (1983, 1987, 1998, 2000) have cre-ated literacy scales to assess lay science knowledge that have been appliedlocally, regionally, nationally, and internationally. The work of the NationalScience Foundation’s Science and Engineering Indicators (National ScienceBoard 2000) enables researchers to compare how much students know aboutscience throughout the United States and around the world and to providebenchmarks to assess improvements (or declines) over time. Efforts such asthe Science and Engineering Indicators provide a sophisticated,postexposure measure of the impact of science news, which is consistent withthe classic model’s rationales to encourage science communication forcitizens.

The other branch of audience-based research in the scientific literacy tra-dition explores audience predispositions. Prewitt (1982, 1983) found anaudience segment (based on their knowledge, attitudes, habits, and interests)was predisposed to read, listen, and watch science news, and this audiencesegment was termed an “attentive public” for science. By studying public


attitudes in conjunction with demographic variables, such as income, educa-tion, social attitudes, gender, and socioeconomic status, Prewitt (1982, 1983)and Miller (1986, 1998, 2000) roughly identified audiences that were proneor unlikely to be interested in science news.

Gerbner (Gerbner et al. 1980, 1981, 1986) also described an “inadvertentaudience” that is neither attentive nor inattentive to science, health news, orother news. To Gerbner (1987; see also Gerbner et al. 1980, 1981, 1986),inadvertent audiences often heavily scan news and popular culture content ontelevision and form impressions about surrounding society and lifestyle hab-its by casual and constant exposure to media content. Gerbner (Gerbner et al.1980, 1981, 1986) argued many Americans believe the prevalence of crime isworse than actual statistics because crime news is reported frequently withinthe news and mass media without an accompanying statistical context. Tele-vision viewers also sometimes ignore evidence about the risks ofhealth-related issues, such as smoking, alcohol consumption, and weightcontrol, and gravitate toward unhealthy behaviors that are unrealisticallydepicted in popular television programming (Gerbner et al. 1980, 1981,1986; Signorielli 1993).

Assessment of risk perception further elaborates research about audiencepredispositions by predicting how a range of specific issues fosters citizenoutrage about science and technology (such as whether adults perceive a riskas voluntary, e.g., driving, or involuntary, e.g., living thousands of miles fromthe Chernobyl nuclear power plant) (Slovic 1987a, 1987b; Slovic, Fischoff,and Lichtenstein 1993).

Both cognitive postexposure research and predisposition research repre-sent core contributions to understanding science mass communication pro-cesses in terms of conceptual development and practice. They continue a leg-acy to explore the conceptual purpose of science communication’s flow,propose what impact science’s mass communication has on audience knowl-edge, and help media practitioners and scientists anticipate complex socialresponses.

Some insights from this research genre also led to an important critique ofscience’s public communication by Trachtman (1981) and Burnham (1987)(see also Dornan 1990; Hilgartner 1990). Despite a generation of efforts touse the press to inform citizens about science (based on the classic sciencecommunication model), Trachtman noted scientific literacy in the UnitedStates was declining rather than improving. Trachtman (1981) and Burnham(1987) questioned whether the effort made to elevate the public’s scientificliteracy (by working with the news media) was undermining the desiredeffect. Although neither Trachtman nor Burnham questioned whether theobserved problems lie in the classic science communication conceptual


framework, their work demonstrated that science communication’s classicframework could be turned on its head. Following Trachtman’s andBurnham’s reasoning, the classic model could be used to justify less, notmore, cooperation between scientists, the news media, and the public. In turn,Trachtman (1981) and Burnham (1987) raised fundamental questions lessabout audience research than about the classical model’s conceptual compre-hensiveness. The interest in pursuing alternative conceptual frameworks,which is discussed two sections below, partially may have stemmed fromscholars who acknowledged the poor scientific literacy in the United Statesand a large nonattentive audience for science—but refused to conclude eitherscenario was a rationale to curtail public communication efforts, asTrachtman (1981) and Burnham (1987) suggested.

Other Links in the Scientific Literacy Process:News Sources, News Messages, and Channels

Besides audiences, the classic science communication framework (cou-pled with aforementioned advances in mass communication models) alsogenerated interest in the other links in the process as news travels from scien-tists through journalists to citizens. Westley and MacLean (1957) noted thatcomprehensive research about the mass and news communication processdemanded attention to the receivers of mass communication (the audienceresearch reviewed above) plus news sources, messages, and channels. Duringthe past forty years, Westley and MacLean’s source, message, channel, andreceiver model has been broadened to encompass diverse topics such as theinfluences of a surrounding social climate and culture, prevailing politicaland economic ideologies, social psychology (how persons assess sourcecredibility and are influenced by group and interpersonal pressures), cogni-tive behavioral models (how knowledge is linked to individual actions), andlife skills (how personal behaviors sometimes are influenced by feelingempowered or capable to respond) (Logan and Longo 1999; McQuail 2000).

Singularly within the area of assessing news “messages,” McQuail (2000)found that scholarship has evolved from an emphasis on text to focusing onhow editing decisions cultivate public knowledge and health habits, help setsociety’s public affairs agenda, and indirectly frame how persons mayrespond to the topics discussed within news reports.

Within the broad area that Westley and MacLean (1957) termed“sources,” McQuail (2000) noted that scholarship evolved from evaluatingsource credibility to the complex interactions and professional relationshipamong journalists, news sources, and other participants in the creation of


news, including public information officers. Within the broad area termed“channels,” the Internet rekindled interest in comparing whether newspapers,magazines, radio, television, or the Internet is ideally suited to convey newsand information (McQuail 2000).

As discussed below, the development of science communication scholar-ship regarding source-journalist interactions, how messages reflect newsdecision-making processes, and the comparative capacity of different mediachannels seems to parallel developments within the broader mass communi-cation field. If placed in a conceptual tour under headings “sources,” “mes-sages,” and “channels,” an impressive array of issues regarding science’smass communication have been raised. This section selectively lists some ofthe issues science communication researchers have explored within sources,messages, and channels categories, respectively.

Some Issues Associated withScholarship about News Sources

Within the broad conceptual category that Westley and MacLean (1957)identified as “source,” science communication researchers have explored arange of issues that roughly can be divided into four categories: (1) scientistsas sources and resources, (2) journalists and their role in utilizing sources andresources, (3) public information officials as sources and resources, and (4) thescience policy climate under which scientists, journalists, and public infor-mation officials work.

1. Scientists as Sources and Resources

Within this subcategory, some of the issues raised include the following:

• Who are the sources of science news (Dunwoody 1986)?• How and why does news attention gravitate to similar institutional sources,

such as some major universities and some refereed scientific journals (Nelkin1995)? Is this attention linked to stature among scientific and biomedical pro-fessionals, the infrastructure capacity and willingness of these institutions todeal with the press, a combination of these reasons, or others (Nelkin 1995)?

• What are the norms, conventions, attitudes, and habits of scientists regardingcommunication with the news media and the public? Do attitudes shift as aresult of good or bad experiences with the press and public (Council of Scien-tific Society Presidents 1991; Hart 1984; Laetsch 1987)?

• Is communication with the press or public inimical or helpful to peer evalua-tion among scientists (Dunwoody 1986; Hart 1984)?


• What are some of the characteristics of “visible scientists,” and how does theirparticipation in public communication influence their professional and publicstanding (Dunwoody 1986; Goodell 1977)? In cases of celebrated visible sci-entists, such as Carl Sagan, why does high public standing sometimes conflictwith professional evaluation (Davidson 1999)?

• When is it appropriate for scientists to decline to work with the press and public(e.g., When do national security and corporate trade secrets eclipse publiccooperation?) (Hart 1984; Hartz and Chappell 1997; Laetsch 1987)?

• To what extent are whistleblowers in science organizations, government, andindustry necessary to ensure the full disclosure of science news and informa-tion (Rowan 2001)?

2. Journalists’ Role in UtilizingNews Sources and Resources

Within this subcategory, some of the issues explored include thefollowing:

• What are the attitudes among journalists about the professionalism of scientistsas news sources (Council of Scientific Society Presidents 1991; Krieghbaum1967; Perlman 1974)?

• Do reporters adopt some of the norms and values of scientists when journalistsare heavily dependent on scientists as news sources (Haff 1976; Nelkin 1995;Perlman 1974)?

• Should journalists honor embargoes issued by scientific journals, science orga-nizations, universities, and other news sources (Haff 1976; Krieghbaum 1967;Nelkin 1995)?

• Do news editors (who often are a step removed from reporter-scientist interac-tions) tend to be more critical of science and thereby better reflect public (vs.scientific) perspectives and concerns (Nelkin 1995)?

• To what extent do journalists gravitate toward visible scientists and more visi-ble scientific social institutions (Goodell 1977; Nelkin 1995)?

• To what extent can journalists be manipulated by sources within government,industry, public interest groups, and scientific organizational sources to skewcoverage toward parochial goals (Angell 1996; Nelkin 1995)?

• To what extent can journalists be manipulated by sources outside of govern-ment, industry, public interest groups, and scientific organizational sources,who skew coverage toward parochial goals (Angell 1996; Nelkin 1995)?

• Why is it important for journalists to diversify the range of science sources andresources used in reporting science news (Nelkin 1995)?

• Can journalists and scientists working through a professionally sanctionedinfrastructure (such as the American Association for the Advancement of Sci-ence and the National Association of Science Writers) cooperatively foster


mutual professional understandings and improve the quality of science com-munication to citizens (Hartz and Chappell 1997; Nelkin 1995)?

• Is the quality of science news reporting related to the educational backgroundof a reporter (Hartz and Chappell 1997; Krieghbaum 1967; Perlman 1974)?

• When is it ethical for a reporter to use anonymous sources to obtain controver-sial science information and pursue means such as going undercover to obtainscience news stories (Lambeth 1992)?

3. Public Information Officers andTheir Role as Sources and Resources


• What are the professional roles and functions of public information officers(who work for science and biomedical corporations, public interest organiza-tions, scientific societies, and government) (Rogers 1986, 1997)?

• How do public information officers sometimes work through competing loyal-ties to the public versus their employer or client? Is the role of a science publicinformation officer to maximize full public disclosure of science news andinformation (Rogers 1986, 1997; Salisbury 1997)?

• To what extent do public information officers influence the process of sci-ence’s mass communication by translating science to lay audiences (oftenthrough press releases), bringing stories to journalists’attention, and serving asa liaison between scientists, a science-based organization, and science report-ers (Nelkin 1995; Rogers 1986)?

4. The Science Policy Climate under Which AllSources Work (Scientists, Government and CorporateOfficials, Journalists, and Public Affairs Officers)


• Is the agenda of the science topics that are raised as important issues in publicaffairs a function of competition and negotiation among important socialactors, including government agencies, politicians, corporations, public inter-est groups, scientific societies, and the news media (Hilgartner and Bosk1988)?

• To what extent do government agencies, politicians, corporations, public inter-est groups, and scientific societies dominate the agenda of what science topicsare raised as important issues in public affairs (Logan, Zengjun, and FraserWilson 2000b)?


• Is the tacit influence of major social actors in science policy unquestioned bythe news media, and does this reflect a complicity between the press and pow-erful social institutions (Hardt 1999; Hardt and Carey 2001; Illich 1975)?

• What is the obligation of journalists to lobby for improved access to scientists,science resources, and freedom of information (Hartz and Chappell 1997;Krieghbaum 1967)?

Some Issues Associated withScholarship about Messages

Within the broad conceptual category that Westley and MacLean (1957)identified as a “message,” science communication researchers have exploreda range of issues that Nelkin (1995) divided into “news reporting,” “newsediting,” and “writing decisions.” “News reporting” roughly refers to howindividual journalists make reporting decisions and their impact on newsaccuracy and comprehensiveness. “News editing” refers to how groups ofjournalists within news organizations initially decide what is news (what isselected to broadcast or publish) and its impact on providing a comprehen-sive range of science stories. “News writing” refers to writing motifs and nar-ratives (that inevitably are embedded within news stories) and the formationof public impressions about story topics (Logan, Zengjun, and Fraser Wilson2000b, 6). Within each of these subcategories, some of the issues exploredinclude the following:

1. News Reporting

• Is science reporting accurate and impartial (Singer 1990)?• Does news reporting exaggerate the importance of scientific findings (Logan,

Zengjun, and Fraser Wilson 2000b; Nelkin 1995)?• Does news reporting provide qualifications, caveats, and time frames to

explain issues, such as the timetable for the public availability of scientific dis-coveries, procedures, products, or technology (Wilkins 1987, 1989; Wilkinsand Patterson 1987)?

• To what extent does news reporting provide a social, economic, historical, cul-tural, and scientific context (Friedman 1999; Friedman, Gorney, and Egolf1992; Friedman et al. 1996; Logan, Zengjun, and Fraser Wilson 2000b; Nelkin1995)?

• To what extent does news reporting reflect a correct use of statistics and mathe-matics (Cohn 1988; Paulos 1988, 1995)?

• To what extent does news reporting explain the uncertainty that normallyunderlies scientific findings (Dunwoody 1999; Fumento 1993)?


• To what extent does news reporting help readers distinguish between textbookscience (sophisticated, well-understood scientific areas) and frontier science(areas where research findings are preliminary) (Dunwoody 1999; Rowan1999)?

• Why are science news reports frequently tied to events, such as press releases,speeches, journal article releases, and science convention papers (Boorstin1961; Nelkin 1995)?

• To what extent does the skew toward reporting events result in news with lesshistorical, economic, and educational context (Boorstin 1961; Logan,Zengjun, and Fraser Wilson 2000b; Nelkin 1995)?

• Is science reporting too uncritical about science (Greenberg 1974; Nelkin1991; Perlman 1974)?

• To what extent is reporting about female scientists approached as a featurestory about a person while reporting on male scientists is focused on their work(Blakeslee 1986)?

2. News Editing (or Topic Selection Processes)

• To what extent do news editors believe they must publish or broadcast a scienceor medical news story that is already reported by a competitive news organiza-tion (Shoemaker and Reese 1996)?

• To what extent does news selection among a few major news organizations(e.g., the New York Times, Washington Post, and Associated Press) set theagenda for the science and medical coverage across the United States (Shoe-maker and Reese 1996)?

• To what extent does science and medical news selection gravitate towardapplied as opposed to basic science topics (Hartz and Chappell 1997; Nelkin1995)?

• In health reporting, to what extent do news selection tendencies skew towardmajor diseases (such as heart disease, stroke, and cancer) at the expense ofother health care issues (Cohn 1988)?

• To what extent is public attention about serious public health issues, such asAIDS and smoking risks, associated with the press’attention to covering theseareas (Sontag 1988; Warner 1989)?

• In science reporting, why does the menu of stories skew toward coverage of“big science” projects, such as the space program and genome research(Nelkin 1995)?

• Why do news organizations infrequently provide mobilizing and coping infor-mation or give readers, listeners, and viewers contact information to find outmore about the topics within a news story (Wilkins 1987, 1989)?

• Are news selection processes (e.g., internal newsroom decisions about whatscience and medical topics to cover and avoid) associated with what citizens


believe are important to unimportant public affairs topics (Logan, Fears, andWilson 1997; Mazur 1981)?

• Do these news selection processes (often called agenda setting) result in howpublic affairs priorities are established by politicians and how public funds arespent (or are withheld) for scientific and biomedical research (Hartz andChappell 1997)?

3. News Writing

• To what extent is science news placed in rhetorical contexts, or “frames”(Einsiedel 1992; Logan, Zengjun, and Fraser Wilson 2000a)? For example,how frequently is the underlying topic of medical news stories based on themesthat suggest “new hope” and “no hope” for patients (Cohn 1988)? How fre-quently is environmental reporting framed as a trade-off between protectingnature versus employment opportunities for citizens (Efron 1985; Logan,Fears, and Wilson 1997)?

• Do the impressions left by how news is framed create prevailing impressionsthat foster public attitudes and judgments regarding science policy (Logan,Zengjun, and Fraser Wilson 2000a; Murray, Schwartz, and Lichter 2001)?

• How can rhetorical and other writing techniques improve the public under-standing of science (Rowan 1999)?

Some Issues Associated withScholarship about Channels

While channels have received less attention than the other areas Westleyand MacLean (1957) identified, science communication researchers haveexplored some channel-derived topics, including the following:

• Are magazines and the print media inherently better suited to provide in-depthreporting about science (Freimuth et al. 1984; Haff 1976)?

• Is television better suited to provide a broader acquaintance with science topicsand generate interest in learning more about science topics (Hartz andChappell 1997)?

• To what extent does visualizing complex scientific process (in television orprint) enhance public understanding (Flatow 1997; Ropeik 1997; Rowan1999)?

• To what extent can the interactive, print, and visual capacities of the Internet (asa mass media channel) be optimized to the Internet’s full potential as a scienceeducational tool (Tremayne and Dunwoody 2001)?

• How do nonmedia channels of science information, such as museums, exhibi-tions, and the arts, enhance public understanding? To what extent do nonnews


sources of science information supplement public education and the press assources of science news, information, and socialization (Gregory and Miller1998; Lewenstein 1992)?

Origins of—and the Transition to—the Interactive Science Model

While the lists above are not comprehensive, Weigold’s (2001) literaturereview underscores how many issues within the field can be categorizedwithin Westley and MacLean’s (1957) conceptual headings. Essentially,most science communication research has revolved around (1) the sources ofscience news; (2) how news is reported, edited, and written; (3) the appropri-ate media channel to communicate science; and (4) the audience for science.Weigold finds science communication researchers have been innovators inall four ideas, and as noted above, the foundations of mass communicationresearch can be traced to the initial scientific literacy model. In addition,Weigold notes most of the research radiates a concern about exploring howscience communication is functional or dysfunctional and often providessuggestions to improve the science communication process.

From a conceptual perspective, a generation of research about sources,messages, channels, and receivers unquestionably has advanced both theclassic science communication model and long-standing traditions withinmass communication research. Researchers past and present can be proud ofand continue the legacy of the scientific literacy tradition.

Yet, as the range of understanding of news sources, news messages, chan-nels, and audiences unfolded throughout the twentieth century, importantnew questions were raised, such as why there was a lack of progress in elevat-ing scientific literacy in the United States and why significant audiencesremained inattentive, apathetic, disinterested, or neglected or rejected sci-ence (Dornan 1990; Hartz and Chappell 1997; Hilgartner 1990; Logan 1985;Yankelovich 1982). As Trachtman (1981), Burnham (1987), and later Hartzand Chappell (1997) noted, a generation of sophisticated efforts to work withthe news media to boost awareness, interest, and education about science didnot expansively elevate the nation’s scientific literacy, encourage young per-sons to pursue careers in science, foster interest in increasing public spendingon science and technology research, or create more goodwill toward science.For some citizens, the popularization of science through the news media didnot automatically generate interest or increase popular support for tax-sup-ported research. In addition, for some audiences the idealized, transformativeexperience that Slossen, Heyl, Millikan, and Hale envisioned fell far short of


expectations. As Holton (1993) noted, the growth of “anti-science” perspec-tives in serious scholarship, public opinion, and popular culture was difficultfor scientists to ignore during the last third of the twentieth century.

Following the logic of the prevailing model, some suggested responses toimprove the communication of science to the public in the 1980s and 1990swere to (1) redouble efforts to enlist more journalists and scientists toimprove public understanding of science along classic traditions, as Hartzand Chappell (1997) maintained; (2) admit unmet expectations and reduceinterprofessional efforts to communicate science to lay audiences, asTrachtman (1981) and Burnham (1987) implied; or (3) compromise by tar-geting news and information toward attentive publics for science, as Prewitt(1982, 1983) suggested. While Weigold (2001) asks if science communica-tion research is currently at conceptual crossroads for these reasons, evidencesuggests a quiet shift in the literature probably started in the early 1980s.Regardless of approach, by the early 1980s some scholars realized that freshsolutions were needed to expand the audience for science; generate increasedpublic attention to, interest in, and support for science; and improve citizens’capacity to make better science policy decisions (Dornan 1990; Hilgartner1990; Yankelovich 1982). This article takes the position that the leaders whocriticized the scientific literacy model’s inertia about twenty years ago helpeddevelop the interactive science tradition as a conceptual alternative to the sci-entific literacy model. Their initial criticisms of the scientific literacy modelcentered on the ethics and logic of suggestions to curtail sixty years ofinterprofessional commitment to the lay public, or target most attentive audi-ences. As Yankelovich (1982) implied, it seemed conceptually backward tosuggest that curbing efforts to improve lay understanding of science couldhelp adults or children better function as citizens. The suggestion to curb pub-lic communication efforts also was seen as antithetical to the most founda-tional ethical value in journalism—the public’s right to know (Lambeth1992). While ethical theory defends occasional exceptions to informingaudiences, such as national security and privacy, the inconvenience caused bydeclining scientific literacy did not seem to be an ethically based rationale tocurb communication (Logan 1985). In addition, the well-meaning sugges-tions to persevere along established conceptual and practical lines were notseen as innovations (Logan 1999; Yankelovich 1982).

In turn, the frustrations with a lack of progress and contemporary reme-dies turned into a discussion about science communication’s future and itsconceptual foundation. Without immediate solutions, some researchersbegan to suggest that science communication’s classic conceptual underpin-nings might be insufficiently holistic to account for the social dynamics thatscientists, journalists, communicators, and citizens faced (Rakow 1989;


Yankelovich 1982). In lieu of traditional options, some scholars began tofocus on a new conceptual tradition that could underlie—not replace—thetraditional model of science mass communication (Logan 1991; Yankelovich1982). They hoped a fresh conceptual approach might (1) explain why con-structive news efforts sometimes fostered unreceptive to unappreciativeaudiences, plus (2) provide a fresh start to revise and revive the enthusiasmabout bringing science to lay audiences.

Interestingly, a similar crossroads and a conceptual evolution emergedwithin health communication at the same time. Even after a generation ofadvancement in understanding audiences, messages, and channels and elab-orating social learning theory, Salmon (1989) and Strasburger (1989) notedthat well-planned public health campaigns were only modestly successful inchanging audience health behaviors. As sophisticated as health communica-tion models became, Salmon argued they were less successful in conceptu-ally explaining a range of negative reactions among targeted audiences.Although researchers reported modest gains in audience interest, awareness,and inclinations to change behavior, by the early 1990s, Salmon arguedstate-of-the art campaigns also fostered a simultaneous ill will toward thesocial institutions that generated campaigns and a perception among sometarget audiences that health communication efforts were paternalistic. Asother researchers observed a decade later, “The ethos of intervention cam-paigns reflects tacit assumptions that are resented and rejected by the veryaudiences public health officials hope to reach” (Logan and Longo 1999, 83).

If the structure of health communication campaigns was inimical to theirsuccess, some health communication scholars reasoned, “there was a press-ing need for broader theoretical concepts regarding how health communica-tion occurs” (Logan and Longo 1999, 81). Among health communicationresearchers in the 1990s, both practical and conceptual needs spawned inter-est in an alternative model that could explain consumer resistance and offerfresh strategies to improve lay communication efforts. Simultaneously, tosome science communication researchers, unenthusiastic or inattentive pub-lic responses to popularization efforts spawned a parallel interest in an alter-native model that could conceptually explain public apathy and regenerateinterest in bringing science to citizens.

Although the motivations were slightly different, within health and sci-ence communication research, the development of what Einsiedel andThorne (1999) described as the interactive science model was a response topractical issues and an underlying conceptual paradox.

By the early 1990s, the fields of health, science, and risk communicationconverged on a similar interest: is it possible that the prior conceptual empha-sis on public information via a news or media transmission model might be


undergirded with a fresh focus? All three disciplines also converged on a sim-ilar insight: is it possible to build public communication on a new foundationwhere the structure of public communication rekindles interest among abroader range of citizens in science?

The Interactive Science Tradition

Einsiedel and Thorne (1999) differentiated the interactive tradition fromthe scientific literacy tradition by explaining the interactive tradition under-girds a linear, top-down transmission model with a conceptual alternative.Within the interactive tradition, science knowledge is seen as less fixed (orcertain) and does not necessarily flow from scientific experts through thenews media to citizens. In lieu of pedagogy, the interactive tradition con-ceives mass communication as more of an informal conversation—a sharedand multidirectional experience. The emphasis is less on informing personsthan on improving communication among citizens, scientists, politicians,government and corporate officials, and journalists. The efforts to reestablisha dialogue among citizens, scientists, politicians, government and corporateofficials, and journalists are seen as a vital first step to rekindling publicengagement and interest in science (Yankelovich 1991, 1999).

The interactive tradition’s intellectual roots are derived partially frompolitical science (Putnam 1993, 2000), mass communication (Carey 1989;Rosen 1999), and public opinion research (Yankelovich 1991, 1999). Putnam(1993, 2000) noted how, from the 1960s through the 1990s, declines in vot-ing, community volunteer activity, and perceived credibility of social institu-tions and major professions (among other examples) collectively reflected anerosion in the United States’ social capital. The term “social capital” roughlyrefers to the degree that citizens believe social institutions and the major pro-fessions are responsive to public concerns and are dedicated to improving thequality of life for all citizens. “Social capital” additionally means the degreeof perceived public trust and goodwill toward social institutions as well asother social indicators, such as the public’s confidence that the nation’s eco-nomic, political, and cultural future will be better than its past. Putnamexplained that social capital is less an empirical construct than an informalcultural index that rises and falls over time. Moreover, Putnam argued socialcapital can be elevated by sincere efforts among social institutions to encour-age citizens to participate in civic processes, such as voting, community vol-unteer work, and discussions about public policy issues.

Similar to Putnam (1993, 2000), Yankelovich (1991) found declines in theperceived credibility of social institutions and the major professions (in-


cluding science, medicine, and journalism) were symptoms of citizen alien-ation and anomie. Yankelovich (1991) argued secular alienation was a func-tion of declining citizen engagement in public affairs that in turn he partiallylinked to the top-down, pedagogical process of mass communication.Yankelovich (1991) argued that ironically, by placing emphasis onexpert-to-citizen pedagogy, science communicators inadvertently fosteredpublic alienation, inattentiveness, and disinterest and accidentally cultivatedill will toward science as a social institution.

Yankelovich (1991) emphasized that public participation is eclipsed whencitizens do not have the media and publicity skills or access to enter publicaffairs arenas with capacity or influence parallel to organized social institu-tions (Logan and Longo 1999). Hilgartner and Bosk (1988) added that“expert arenas” (such as the leadership within legislatures, political parties,government, unions, industry, and public interest groups) dominate thedebate about science policy as well as risk and health issues. Expert arenasfocus public attention around parochial concerns, which “influence legisla-tion and spending [and] helps transform popular skepticism into cynicismabout the evidence, motives and credibility of social institutions” (Logan andLongo 1999, 86).

Yankelovich’s (1991) remedy was to focus on the roots of public alien-ation and encourage social institutions and major professions to establishunprecedented efforts for dialogue with citizens. To Yankelovich, the processof seeking dialogue was more than providing an opportunity for public inputinto governmental hearings, holding corporate open houses, or providing let-ters to the editor and discussion groups for Internet readers. A Yankelovich-inspired dialogue is an ongoing, live interaction between scientific experts,policymakers, scientists, lobbyists, and representative citizens regarding themoral, ethical, and affective dimensions of science and medical issues; thelinkage of related epidemiological and toxicological issues to public confu-sion about risk (perception); and a discussion about how power and authorityare advanced to make public policy health decisions (Logan and Longo1999). The purpose is to mediate and inform how biomedicine or science“infuses cultural outlooks, creates options for public consideration, alters theattractiveness of health alternatives, identifies the consequences of publicchoices, helps raise issues to public attention, and influences social valuesand valuation processes” (Logan and Longo 1999, 87).

In the history of mass communication research, the conceptual divisionbetween the interactive and scientific literacy traditions plus applied strate-gies, such as public dialogue, were probably first broadly advanced by Carey(1989). To Carey, pedagogically oriented mass communication efforts (suchas the classic scientific literacy model) reflected an incomplete conceptual


understanding of the processes of social learning, public education, and fos-tering responsive citizenship.

Facts aren’t enough; people look for a sense of authenticity from informationand individuals. The story must “ring true”—reflecting to people a sense ofreality that resonates with their experiences and the general belief that they arebeing squared with. Citizens detect and dislike the inflated language of experts.(Warhover 2000, 50)

While Carey understood that traditional pedagogical models were vital topublic education, he maintained citizen involvement in public affairsfoundationally depended on social rituals, such as New England town meet-ings, where citizens could see their interest in public life result in real deci-sions and social change. As Warhover (2000) explained, “people want a senseof possibility for action before they will get engaged (in social learning andpublic affairs). They need to believe that progress can be made, that they canparticipate effectively” (p. 50).

Inspired partially by Carey’s (1989) and Yankelovich’s (1991) ideas,Rosen (1999) and others (Eksterowicz 2000) encouraged news organizationsto foster lay participation in civic processes. Rosen called these pioneerefforts civic or public journalism, which supplements traditional newsreporting with specialized coverage that (among other issues) fosters moredialogue between experts and citizens. In the past decade, an array of newsorganizations throughout the United States has adopted civic or public jour-nalism strategies (Eksterowicz 2000). While the underlying ideas and strate-gies surrounding civic or public journalism are controversial among journal-ists (Woo 2000), their existence is seen as a historic, conceptual shift in howjournalists perceive their traditional roles and functions (Eksterowicz 2000;Rosen 1999).

Returning to its conceptual impact within science communication, theinteractive tradition starts with fundamentally different questions than doesthe scientific literacy legacy. The scientific literacy tradition conceptuallyfocuses on how accuracy and context are maintained as blocks of knowledgeand migrate from scientific experts through media channels to citizens. In theinteractive tradition, on the other hand, the foundational questions includethe following:

a. How is learning about science cultivated when the audience is inadvertent, dis-affected, alienated, or unmotivated?

b. How is learning about science fostered when the process of social learningsometimes is not linear and top-down?


c. How is learning about science fostered when, sometimes, adults and youngpersons perceive science communication efforts as didactic or paternalistic?

d. How can science, the news media, and other major social institutions andmajor professions encourage participation in a democratic society?

e. How can science, the news media, and other major social institutions andmajor professions better connect citizens to civic processes?

f. How is credibility, goodwill, and trust in science (and other social institutions)reestablished once traditional strategies to inform citizens fall short ofexpectations?

While the conceptual history of the interactive tradition reflects a rich leg-acy, its range of practical applications and formal evaluation are in their for-mative stages. Some of the interactive science model’s concepts and strate-gies only recently emerged as important topics within risk communicationresearch (Plough and Krimsky 1987; Powell and Leiss 1997). In health com-munication, the interactive tradition is reflected by a recent interest in admin-istering town meetings as an intervention to generate audience participationand interest. The town meeting is conceived as an initial step to establish whatlater becomes a traditional campaign of providing health messages to targetaudiences (Fawcett et al. 2000; Green and Kreuter 1999; Guttman 2000). Oneof the first comprehensive, empirical assessments of the community impactof civic journalism found the effort modestly increased citizen involvementin public affairs, modestly elevated interest in learning more about socialissues, and boosted the perceived credibility of participating social institu-tions, including the news media (Lambeth, Meyer, and Thorson 1998).

Nevertheless, the applications and evaluations of the interactive tradition(in science and health communication) are not extensive, especially in com-parison with the wide-ranging research that underpins the classic sciencecommunication model. In addition, it is premature to assert that the concep-tual ideas within the interactive model are grounded by research findings. It isyet to be determined if strategies derived from the interactive tradition (suchas social dialogue) have a socially desirable impact (such as elevating socialcapital).

On the other hand, the interactive model represents an important contribu-tion to the conceptual history of science communication because it providesalternative perspectives and pragmatic strategies that revitalize how publiccommunication might be approached. If nothing else, the interactive sciencetradition challenged a conceptual inertia that emerged within the field abouttwenty years ago. The interactive tradition created energy and enthusiasm toexpand public involvement in science, refocused professional interests onwhat practitioners can do to elevate public life, and rekindled the field of


science communication’s conceptual momentum. It also provided a freshsense of direction without undermining the efforts of those who sought toimprove science communication in traditional ways.

The Two Traditions and Future Opportunities

While the interactive and scientific literacy traditions are different, theyare not mutually exclusive. Although the interactive tradition responds toconceptual binds within the scientific literacy model, the intent of the interac-tive tradition is to underlie—rather than replace—the traditional view of thescience communication process. The interactive tradition does not quarrelwith the idea that citizens should be better informed about science, nor does itoverlook the important roles scientists, journalists, public information offi-cers, public interest groups, corporations, governmental agencies,nongovernmental agencies, and other professionals play in providinghigh-quality science information to the public.

To put this another way, the field of science communication is conceptu-ally expanded—not confounded—by the existence of two different concep-tual traditions. The interactive science tradition may provide a more compre-hensive explanation of how public communication processes occur, but itdoes not conceptually threaten traditional goals of informing adults and chil-dren about science. The interactive tradition is uncritical of the thousands ofwell-intentioned scientists, journalists, public information officers, and oth-ers who try to translate science into useful and understandable narratives forcitizens. The interactive tradition simply provides some new issues for sci-ence communication practitioners to consider and a range of fresh strategiesto attempt to supplement traditional approaches.

In fact, science communication is fortunate to have two conceptual tradi-tions that provide a range of options for researchers and practitioners. Whilethe emphasis above was on the differences between the interactive scienceand scientific literacy models, a review of their development also reveals thatthere are commonalities between science communication’s subdisciplinesand common bases for opportunities and cooperation. Although journalistscurrently are organized into different peer organizations depending onwhether they cover science, health, medicine, environment, technology,nutrition, or agriculture—and the broad field of mass communication of sci-ence is spread across three disciplines (health, science, and risk communica-tion)—two identifiable conceptual traditions link all science communicators.The scientific literacy and interactive traditions provide a conceptual basis


for disparate practitioner and scholarly groups to compare professionalissues, discuss research findings, evaluate their fields on a cross-disciplinarybasis, and learn from each other’s work. The commonalities also provide abasis for scientists to discuss the purpose of public communication with jour-nalists and furnish a basis for outreach and dialogue with vital actors in thescience communication process, such as public information officials whowork with government, corporations, public interest groups, and politicians.

While there are scholars who cross the boundaries of risk and sciencecommunication or health and science communication, unfortunately, there islittle sustained effort within the three disciplines for researchers to meet, dis-cuss their work, and seek interdisciplinary research opportunities. Similarly,leaders within the National Association of Science Writers, the Society ofEnvironmental Journalists, and the Association of Health Care Journalistsrarely collaborate on workshops, common skills, Web sites, and other profes-sional development activities.

While recent conceptual history suggests social and professional learningis rarely linear and progress cannot be rushed or contrived, it is straightfor-ward in suggesting that more interprofessional and multidisciplinary effortsto discuss common issues might accelerate the pace at which science journal-ism and science communication progress. The overlaps in conceptual tradi-tions in science, risk, and health communication, and between all sciencecommunication practitioners, are a foundation for collaboration. The com-mon roots, objectives, challenges, transformations, and questions remain acommon point of pride and an underutilized foundation for progress.

References

Angell, M. 1996. Science on trial: The clash of medical evidence. New York: Norton.Bandura, A. 1977. Social learning theory. Englewood Cliffs, NJ: Prentice Hall.Blakeslee, S., ed. 1986. Human heart replacement: A new challenge for physicians and report-

ers. Los Angeles: Foundation for American Communications.Boorstin, D. 1961. The image: A guide to pseudo events in America. New York: Athaneum.Bronowski, J. 1973. The ascent of man. Boston: Little, Brown.Burnham, J. 1987. How superstition won and science lost: Popularizing science and health in

the United States. New Brunswick, NJ: Rutgers University Press.Carey, J. W. 1989. Communication as culture: Essays on media and society. Boston: Unwin

Hyman.Cohn, V. 1988. News and numbers: A guide to reporting statistical claims and controversies in

health and other fields. Ames: Iowa State University Press.Council of Scientific Society Presidents. 1991. Bridging the communications gap: A

Wingspread workshop on strengthening relationships between scientists and journalists.Washington, DC: Council of Scientific Society Presidents.


Davidson, K. 1999. Carl Sagan: A life. New York: John Wiley.Dornan, C. 1990. Some problems of conceptualizing the issue of “science and the media.” Criti-

cal Studies in Mass Communication 7:48-71.Dunwoody, S. 1986. The scientist as source. In Scientists and journalists: Reporting science as

news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers, 3-16. New York: FreePress.

. 1999. Scientists, journalists and the meaning of uncertainty. In Communicating uncer-tainty: Media coverage of new and controversial science, edited by S. M. Friedman,S. Dunwoody, and C. L. Rogers, 59-80. Mahwah, NJ: Lawrence Erlbaum.

Efron, E. 1985. The apocalyptics: How environmental politics controls what we know about can-cer. New York: Simon & Schuster.

Einsiedel, E. F. 1992. Framing science and technology in the Canadian press. Public Under-standing of Science 1:164-88.

Einsiedel, E. F., and B. Thorne. 1999. Public responses to uncertainty. In Communicating uncer-tainty: Media coverage of new and controversial science, edited by S. M. Friedman,S. Dunwoody, and C. L. Rogers, 43-58. Mahwah, NJ: Lawrence Erlbaum.

Eksterowicz, A. J. 2000. The history and development of public journalism. In Public journal-ism and political knowledge, edited by A. J. Eksterowicz and R. N. Roberts, 3-20. Lanham,MD: Rowman & Littlefield.

Farquhar, J. W., P. Magnus, and N. Maccoby. 1981. The role of public information and educationin cigarette smoking control. Canadian Journal of Public Health 72:412-60.

Fawcett, S. B., V. T. Francisco, A. Paine-Andrews, and J. A. Schultz. 2000. A model memoran-dum of collaboration: A proposal. Public Health Reports 115:174-79.

Flatow, I. 1997. Broadcast science journalism: Magazine style. In A field guide for science writ-ers, edited by D. Blum and M. Knudson, 38-42. New York: Oxford University Press.

Freimuth, V., R. Greenburg, J. DeWitt, and R. Romano. 1984. Covering cancer: Newspapers andthe public interest. Journal of Communication 34:62-73.

Friedman, S. M. 1999. The never ending story of dioxin. In Communicating uncertainty: Mediacoverage of new and controversial science, edited by S. M. Friedman, S. Dunwoody, andC. L. Rogers, 113-36. Mahwah, NJ: Lawrence Erlbaum.

Friedman, S. M., C. M. Gorney, and B. P. Egolf. 1992. Chernobyl coverage: How the US mediatreated the nuclear industry. Public Understanding of Science 1:305-23.

Friedman, S. M., K. Villamil, R. A. Suriano, and B. P. Egolf. 1996. Alar and apples: Newspapers,risk and media responsibility. Public Understanding of Science 5 (1): 1-20.

Fumento, M. 1993. Balancing technology and the environment: Science under siege. New York:William Morrow.

Gerbner, G. 1987. Science on television: How it affects public conceptions. Issues in Science andTechnology 3:109-15.

Gerbner, G., L. Gross, M. Morgan, and N. Signorielli. 1980. The “mainstreaming” of America:Violence profile No. 11. Journal of Communication 30 (3): 1-29.

. 1981. Health and medicine on television. New England Journal of Medicine 305 (15):901-4.

. 1986. Living with television: The dynamics of the cultivation process. In Perspectiveson media effects, edited by J. Bryant and D. Zillman, 17-40. Hillsdale, NJ: LawrenceErlbaum.

Goodell, R. 1977. The visible scientists. Boston: Little, Brown.Green, L. W., and M. W. Kreuter. 1999. Health promotion planning: An educational and ecolog-

ical approach. 3d ed. Mountain View, CA: Mayfield.


Greenberg, D. 1974. Let’s hear it for science. Columbia Journalism Review 13:19-24.Gregory, J., and S. Miller. 1998. Science in public: Communication, culture and credibility. New

York: Plenum.Guttman, N. 2000. Public health communication interventions: Values and ethical dilemmas.

Thousand Oaks, CA: Sage.Haff, G. 1976. Science writing in American mass media. Nieman Reports 30:18-24.Hardt, H., 1999. Shifting paradigms: Decentering the discourse of mass communication. Mass

Communication and Society 3:175-84.Hardt, H., and J. W. Carey. 2001. Social theories of the press. 2d ed. Lanham, MD: Rowman &

Littlefield.Hart, R. 1984. Shamans and criers: Responsibilities in science reporting. Quill 72:24-28.Hartz, J., and R. Chappell. 1997. Worlds apart: How the distance between science and journal-

ism threatens America’s future. Nashville, TN: First Amendment Center, Freedom Forum.Hilgartner, S. 1990. The dominant view of popularization: Conceptual problems, political uses.

Social Studies of Science 20:519-39.Hilgartner, S., and C. Bosk. 1988. The rise and fall of social problems: A public arenas model.

American Journal of Sociology 94:53-78.Holton, G. 1993. Science and anti-science. Cambridge, MA: Harvard University Press.Illich, I. 1975. Medical nemesis: The expropriation of health. London: Calder & Boyers.Krieghbaum, H. 1967. Science and the mass media. New York: New York University Press.Laetsch, W. 1987. A basis for better public understanding of science. In Communicating science

to the public, CIBA Foundation Conference, 1-10. New York: John Wiley.Lambeth, E. 1992. Committed journalism: An ethic for the profession. 2d ed. Bloomington: Indi-

ana University Press.Lambeth, E., P. Meyer, and E. Thorson. 1998. Assessing public journalism. Columbia: Univer-

sity of Missouri Press.Lewenstein, B. 1992. Public understanding of science in the United States after World War II.

Public Understanding of Science 1:45-68.Logan, R. A. 1985. Commentary: Rationales for investigative and explanatory trends in science

reporting. Newspaper Research Journal 7 (1): 53-58.. 1991. Popularization and secularization: Media coverage of health. In Risky business:

Communicating issues of science, risk and public policy, edited by L. Wilkins and P. Patterson,43-60. New York: Greenwood.

. 1999. Worlds apart: A review essay. Science Communication 20:337-43.Logan, R. A., L. Fears, and N. F. Wilson. 1997. Social responsibility in science news: Four case

studies. Washington, DC: Media Institute.Logan, R. A., and D. R. Longo. 1999. Rethinking anti-smoking campaigns: Two generations of

research and some issues for the next. Journal of Health Care Finance 25 (4): 77-90.Logan, R. A., P. Zengjun, and N. Fraser Wilson. 2000a. Prevailing impressions in science and

medical news: A content analysis of the Los Angeles Times and the Washington Post. ScienceCommunication 22:27-45.

. 2000b. Science and medical coverage in the Los Angeles Times and the WashingtonPost: A six year perspective. Science Communication 22:5-26.

MacLean, M. S. Jr. 1965. Some multivariate designs for communication research. JournalismQuarterly 27 (4): 614-22.

. 1967. Systems of news communication. In Communication: Theory and research,edited by L. Thayer, 289-309. Springfield, IL: Charles C Thomas.


Mazur, A. 1981. Media coverage and public opinion on scientific controversies. Journal of Com-munication 31:106-15.

McQuail, D. 2000. McQuail’s mass communication theory. 4th ed. Thousand Oaks, CA: Sage.Miller, J. 1983. Scientific literacy: A conceptual and empirical review. Daedalus 112:29-48.. 1986. Reaching the attentive and interested publics for science. In Scientists and jour-

nalists: Reporting science as news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rog-ers, 55-69. New York: Free Press.

. 1987. Scientific literacy in the United States. In Communicating science to the public,CIBA Foundation Conference, 19-37. New York: John Wiley.

. 1998. The measurement of scientific literacy. Public Understanding of Science 7:1-21.. 2000. The development of civic scientific literacy in the United States. In Science, tech-

nology and society, edited by L. Kumar and D. Chubin, 21-47. New York: Plenum.Murray, D., J. Schwartz, and S. R. Lichter. 2001. It ain’t necessarily so: How media make and

unmake the scientific picture of reality. Lanham, MD: Rowman & Littlefield.National Science Board. 2000. Science and engineering indicators 2000. Arlington, VA:

National Science Foundation.Nelkin, D. 1991. Why is science writing so uncritical of science? In Risky business: Communi-

cating issues of science, risk and public policy, edited by L. Wilkins and P. Patterson, ix-xiii.New York: Greenwood.

. 1995. Selling science: How the press covers science and technology. Rev. ed. New York:Freeman.

Paulos, J. 1988. Innumeracy: Mathematical illiteracy and its consequences. New York: Vintage.. 1995. A mathematician reads the newspaper. New York: Basic Books.Perlman, D. 1974. Science and the mass media. Daedalus 103:207-22.Plough, A., and S. Krimsky. 1987. The emergence of risk communication studies: Social and

political context. Science, Technology, & Human Values 12 (3-4): 4-10.Powell, D., and W. Leiss. 1997. Mad cows and mother’s milk: The perils of poor risk communi-

cation. Montreal, Canada: McGill Queens University Press.Prewitt, K. 1982. The public and science policy. Science, Technology, & Human Values 36:5-14.. 1983. Scientific illiteracy and democratic theory. Daedalus 112:49-64.Putnam, R. D. 1993. Making democracy work. Princeton, NJ: Princeton University Press.. 2000. Bowling alone: The collapse and revival of American community. New York:

Simon & Schuster.Rakow, L. F. 1989. Information and power: Toward a critical theory of information campaigns.

In Information campaigns: Balancing social values and social change, edited by C. Salmon,164-84. Newbury Park, CA: Sage.

Rogers, C. L. 1986. The practitioner in the middle. In Scientists and journalists: Reporting sci-ence as news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers, 42-54. New York:Free Press.

. 1997. Introduction. In A field guide for science writers, edited by D. Blum and M. Knudson,213-16. New York: Oxford University Press.

Ropeik, D. 1997. Broadcast science journalism: Reporting news. In A field guide for sciencewriters, edited by D. Blum and M. Knudson, 35-38. New York: Oxford University Press.

Rosen, J. 1999. What are journalists for? New Haven, CT: Yale University Press.Rowan, K. 1999. Effective explanation of uncertain and complex science. In Communicating

uncertainty: Media coverage of new and controversial science, edited by S. M. Friedman,S. Dunwoody, and C. L. Rogers, 202-23. Mahwah, NJ: Lawrence Erlbaum.


. 2001. Whose side are you on? Perceptions of effective science news coverage amongscientists, journalists, and public affairs officers. Paper delivered to the American Associa-tion for the Advancement of Science annual meeting, San Francisco, February.

Salisbury, D. 1997. College and universities. In A field guide for science writers, edited by D. Blumand M. Knudson, 217-25. New York: Oxford University Press.

Salmon, C. T. 1989. Campaigns for social “improvement”: An overview of values, rationalesand impacts. In Information campaigns: Balancing social values and social change, editedby C. Salmon, 19-53. Newbury Park, CA: Sage.

Severin, W. J., and J. W. Tankard Jr. 1997. Communication theories: Origins, methods and usesin the mass media. New York: Addison-Wesley.

Shannon, C., and W. Weaver. 1949. The mathematical theory of communication. Urbana: Uni-versity of Illinois Press.

Shoemaker, P., and S. Reese. 1996. Mediating the message: Theories of influence on mass con-tent. White Plains, NY: Longman.

Signorielli, N. 1993. Mass media images and impact on health: A sourcebook. Westport, CT:Greenwood.

Silver, B. 1998. The ascent of science. New York: Oxford University Press.Singer, E. 1990. A question of accuracy: How journalists and scientists report research on haz-

ards. Journal of Communication 37:10-36.Slossen, E., and J. E. Downey. 1922. Plots and personalities: A new method of testing and train-

ing the creative imagination. New York: Century.Slovic, P. 1987a. Informing the public about risk. Risk Analysis 4:403-15.. 1987b. Perception of risk. Science 35:280-85.Slovic, P., B. Fischoff, and S. Lichtenstein. 1993. Rating the risks. In Readings in risk, edited by

T. Glickman and M. Gough, 61-75. Washington, DC: Resources for the Future.Snow, C. P. 1993. The two cultures. Cambridge, UK: Cambridge University Press.Sontag, S. 1988. AIDS and its metaphors. New York: Farrar, Strauss and Giroux.Stephenson, W. 1973. Lake Ozark symposium on science news. Columbia, MO: National Sci-

ence Foundation.Strasburger, V. 1989. Prevention of adolescent drug abuse: Why “just say no” just won’t work.

Journal of Pediatrics 114:4, 676-81.Tobey, R. 1971. The American ideology of national science. Pittsburgh, PA: University of Pitts-

burgh Press.Trachtman, L. 1981. The public understanding of science effort: A critique. Science, Technol-

ogy, & Human Values 36:10-15.Tremayne, M. W., and S. Dunwoody. 2001. Using think aloud protocols to compare the useful-

ness of science Web sites to the public. Paper delivered to the American Association for theAdvancement of Science annual meeting, San Francisco, February.

Wallack, L. 1993. Media advocacy and public health: Power for prevention. Newbury Park, CA:Sage.

Warhover, T. A. 2000. Public journalism and the press: The Virginian-Pilot experience. In Publicjournalism and political knowledge, edited by A. J. Eksterowicz and R. N. Roberts, 43-60.Lanham, MD: Rowman & Littlefield.

Warner, K. 1989. The epidemiology of coffin nails. In Health risks and the press, edited byM. Moore, 73-89. Washington, DC: Media Institute.

Weigold, M. F. 2001. Communicating science: A review of the literature. Science Communica-tion 23:164-94.

Westley, B. H., and M. MacLean Jr. 1957. A conceptual model for communication research.Journalism Quarterly 34:1, 31-38.


Wilkins, L. 1987. Shared vulnerability: The mass media and American perception of the Bhopaldisaster. Westport, CT: Greenwood.

. 1989. Bhopal: The politics of mediated risk. In Bad tidings: Communication and catas-trophe, edited by L. M. Walters, L. Wilkins, and T. Walters, 21-34. Hillsdale, NJ: LawrenceErlbaum.

Wilkins, L., and P. Patterson. 1987. Risk analysis and the construction of news. Journal of Com-munication 37:80-92.

Woo, W. 2000. Public journalism: A critique. In Public journalism and political knowledge,edited by A. J. Eksterowicz and R. N. Roberts, 21-42. Lanham, MD: Rowman & Littlefield.

Yankelovich, D. 1982. Changing public attitudes to science and the quality of life. Science, Tech-nology, & Human Values 39:23-29.

. 1991. Coming to public judgment: Making democracy work in a complex world. Syra-cuse, NY: Syracuse University Press.

. 1999. The magic of dialogue: Transforming conflict into cooperation. New York:Simon & Schuster.

ROBERT A. LOGAN is a professor, the associate dean for undergraduate studies, andthe director of the Science Journalism Center at the School of Journalism, University ofMissouri–Columbia.


SCIENCE COMMUNICATIONWeigold / A REVIEW OF THE LITERATUREThis article provides an overview of science communication, which is a vital area of mass com-munication scholarship. The review is organized around the key players, including news organi-zations, reporters, science information professionals, scientists, and audiences. Also reviewed isthe problem of science communication, which may be partly responsible for widespread scienceilliteracy. Ways of improving the practice of science communication and an agenda for futureresearch are offered.

Communicating ScienceA Review of the Literature

MICHAEL F. WEIGOLDUniversity of Florida

Media messages about science have long attracted attention from communi-cation scholars (Cronholm and Sandell 1981; Grunig 1979, 1983; Jerome1986; Lewenstein 1992). Perhaps this is surprising since the attention givenscience in most news media is small in comparison to that accorded to busi-ness, politics, or even sports and entertainment. But, scholars in this areaargue that the importance of science news is poorly benchmarked by theattention it receives in most mass media. In an era of unprecedented techno-logical and scientific advances, many of which have the potential to radicallychange human existence, science news is important.

This article provides a brief overview of science communication scholar-ship by first attempting to demarcate the subject of science, then presentingthe key players (news organizations, journalists, science information profes-sionals, scientists, and audiences) and reviewing research detailing their

Author’s Note: Preparation of this overview was facilitated by a grant from the Marshall SpaceFlight Center, Huntsville, Alabama. Address correspondence to Michael F. Weigold, AssociateProfessor, Department of Advertising, P.O. Box 118400, University of Florida, Gainesville, FL32611-8400; phone: 352-392-8199; fax: 352-846-3015; e-mail: [email protected].


164

interrelationships. In the final section, I present a subjective agenda for sci-ence communication scholarship.

What Is Science?

People use “science” to refer to a broad range of activities. It includes thework of academic scientists seeking knowledge for its own sake (basic sci-ence) and the activities of scientists who explore solutions to immediateproblems and concerns (applied science). A broad definition of sciencemight include technologists who use fundamental knowledge to develop anddesign new products, whereas a narrower definition would exclude thisgroup.

Friedman, Dunwoody, and Rogers (1986) proposed a broad definition:“ ‘science’ comprises not only the biological, life, and physical sciences butalso the social and behavioral sciences and such applied fields as medicine,environmental sciences, technology, and engineering” (p. xv). They addedthat “ ‘science writing’ includes coverage of these fields as well as the politi-cal, economic, and social aspects of science” (p. xv).

Science writers and journalists confront definitions of science when theydecide which activities to monitor and explain. Sharon Begley (cited in Hartzand Chappell 1997) of Newsweek suggested that at “Newsweek, science isbasic research.” She continued:

I cover everything from archeology to genetics, neuroscience, and physics. I donot do medicine, which is defined as anything having to do with sick people.And I don’t do technology. I’ll do genetics. I’ll do neuroscience. But once itgets into somebody sick, I give it to “medicine.” (P. 51)

The New York Times, in its weekly science sections, also distinguishes “sci-ence” from “technology.”

Largely unknown is what exactly audiences consider to be science stories.A story on a proposal to build a nuclear power plant may be viewed as a politi-cal story, a big-business story, or an economic story as easily as a sciencestory. The importance of science or scientists to many science-related storiesmay be quite small (Burnham 1987). Recent coverage of lawsuits over breastimplants and Gulf War diseases gave relatively little attention to scientificefforts aimed at determining whether victims suffered to a greater extent thanmight be expected by chance (in both cases, there was considerable evidencethat they had not). Science was a part of the stories, but just a small part.

Weigold / A REVIEW OF THE LITERATURE 165

News Organizations and Science

News media have historically accorded science great importance. In thenineteenth century, newspapers reprinted lectures by Thomas Huxley, LouisAgassiz, and Asa Gray, and one issue of the New York Tribune published thetext of physics lectures by John Tyndall. During the 1920s, press magnateEdwin W. Scripps launched Science Service, a news agency offering the“drama [that] lurks in every test tube.” Science coverage may have reached itszenith during the Second World War, when science and technology were seenas integral to victory. The launching of Sputnik led to a reevaluation of sci-ence education in the United States and to renewed interest in science gener-ally (Shortland and Gregory 1991).

Modern news organizations are more likely to view science as a nichearea; thus, in larger news organizations science may be covered by a beatreporter while in smaller organizations science reporting is more typicallyhandled by a general assignment reporter or by using wire services (Fried-man 1986). The medium itself also affects the quality and amount of sciencenews. Most in-depth reporting is done by newsmagazines, followed by largenational papers. Wires, small dailies, and broadcast stations are least likely tohave the time or money for in-depth science coverage (Ward 1992).

In addition, science news competes with other kinds of news for a rela-tively small amount of space and time. Friedman (1986) estimates that per-haps 5 percent of a typical newspaper is reserved for news of the day, leadingmost papers to place heavy emphasis on story brevity and simplicity. Cover-age of issues in broadcast reports is even tighter. Because effectively tellingscience stories often requires considerable background information, sciencewriters face a difficult challenge.

Several researchers have raised the gatekeeping question: how does newsabout anything, including news about science, pass through the editorial gateto become content? Shoemaker and Reese (1991) suggested all news organi-zations rely on “craft norms” for generating news. These include prominence/importance (How many lives are affected? Fatalities are “worth” more thanproperty damage. Actions of the powerful are more newsworthy than actionsof ordinary people or the poor), human interest (including the activities ofpeople with no direct impact on an audience member’s life other than thatcreated by their own fame, i.e., celebrities, gossip, human dramas), conflict/controversy (Conflict is presumed to alert audiences to important issues. It isalso believed to be inherently more interesting than harmony), the unusual,timeliness, and proximity (Events that happen nearby are considered morenewsworthy). Research has confirmed that these criteria are relevant for a


newspaper’s decision about science coverage as well (David 1996; Ramsey1989; Singer, Endreny, and Glassman 1991).

Other constraints that influence news selection include the complexity ofdeadlines, the unpredictability of occurrences, and the news organization’sability to adapt to physical limits, including limits of time and space (Lieblerand Bendix 1997). Reporters rely on routines that provide access to news,such as press conferences, announcements, and scientific meetings. Becauseof limits of time and resources, reporters often work from “predefinedangles” or frames that provide themes around which to build stories (Baker1986; Shoemaker and Reese 1991).

News organizations also rely heavily on each other for ideas. Gans (1979)argued that editors read elite media such as the New York Times and Washing-ton Post for story ideas, eliminating the need for an independent judgment ofnewsworthiness, a function described by media scholars as “inter-media”agenda setting (Breed 1980; Shoemaker and Reese 1991).

Modern coverage of science varies considerably within and across media.Larger newspapers with better educated readers, such as the New York Times,“cater to an audience interested in reading about some advances in science ormedicine that will be ignored by the editors of the New York Daily News”(Burkett 1986, 12). Newspapers that carry regular science sections as com-pared to those that do not also give greater coverage to science in the newssection (Bader 1990), particularly for stories about basic research. Televisionnews, with its small newshole, often squeezes coverage of science to a bareminimum (Altheide 1976).

Competing media may emphasize different aspects of the same story. Forexample, elite British newspapers emphasize credible sources and scienceprofessionals, whereas popular papers focus more on the consumer’s per-spective (Entwistle and Hancock-Beaulieu 1992). Evans et al. (1990) com-pared the coverage of science stories between elite and tabloid Americannewspapers and found that compared to the tabloids, elite media providemore details about findings and methods employed in the research.

News organizations must also make decisions about which science topicsto cover (Hilgartner and Bosk 1988). Dennis and McCartney (1979) foundthat science writers at large newspapers favored stories about medicine, theenvironment, and technology over stories about the physical and behavioralsciences. And, coverage of scientific ideas is often a function of some news-worthy event rather than the ideas themselves. For example, Caudill (1987,1989) found that coverage of evolution was less a function of new scientificfindings than of anniversaries such as Darwin’s death, the centennial of hisbirth, and the Scopes Monkey Trial. On the other hand, coverage of some


issues, such as AIDS, seems less linked to concrete events and other tradi-tional determinants of newsworthiness (Grube and Boehme-Duerr 1988).

Science is also covered in business, trade, or industrial publications thatfocus on the interests of managers or stockholders in major companies whorequire more than popularization. For example, while some science writerswork as journalists, others work for companies and institutions, “producingreports for a wide range of purposes. There may be press releases promoting acompany product, a brochure explaining a process in layman’s language, or amagazine for stockholders or employees” (Burkett 1986, 13).

Television creates images of science that in turn have implications for howscience is viewed and understood. These images can come from science-related programming, including public television programs such as NOVA,commercial programs such as National Geographic, and the content of somecable networks, such as the Discovery Channel and Animal Planet. Suchimages also can come from programming that is not explicitly concernedwith science but in which science plays a dramatic role, such as science fic-tion programs (Gerbner 1987). Surveying the attitudes and knowledge ofconsumers of science fiction programming suggests that they constituteanother important public of science communication (Banks and Tankel1990). A number of years ago, fans of the science fiction program Star Treklobbied heavily (and successfully) to have NASA name one of the space shut-tles Enterprise (see http://www.pao.ksc.nasa.gov/kscpao/shuttle/resources/orbiters/enterprise.html).

Specialty magazines offer some of the richest and most sophisticated cov-erage of science for general audiences. Scientific American employs editors,but it is scientists, not reporters, who write the stories. Less knowledgeablereaders who might have difficulty with Scientific American can still satisfytheir curiosity with “popularized” magazines such as Discover and PopularScience (Burkett 1986).

Although largely ignored in mass communication scholarship, generalaudience science books may play an important function in the popularizationof science. Such books may represent the public’s only sophisticated encoun-ters with physics (Gleick 1987; Hawking 1988), evolution (Wright 1994),language (Pinker 1994), astronomy (Ferris 1997; Sagan 1980), natural his-tory (Gould 1995), geography (National Geographic Society 1976), mathe-matics (Paulos 1988), or scientists (Boorstin 1983). The popularity and prev-alence of excellent books on science topics suggest that there is an audienceinterested in science issues. Future scholarship is needed to answer some im-portant questions, such as, Who are the readers of these books? What are theylearning? What is the quality of science in such books? What are the opinions


of readers of science books about science policy? and Are science book read-ers interested in specialty areas of science or in science generally?

Scholars are also just beginning to explore the impact of the World WideWeb on communicating science. Space does not permit a detailed explorationof this new medium, but it seems clear that it has the potential to dramaticallychange the relationships of the players in science communication. This is sofor at least four important reasons. First, the Web permits scientists and theirorganizations to communicate directly with audiences. The mediation ofnews organizations is no longer a necessity. Second, the Web largely elimi-nates the severe space and time restrictions inherent in ordinary news media.It therefore allows for complex, sophisticated, and interconnected pieces ofinformation. Third, the Web combines the information richness of print withthe demonstration power of broadcast in a seamless, accessible, interactivefashion. Finally, the Web is an instantaneous two-way communicationsmedium, allowing one-to-one, one-to-many, many-to-one, and many-to-many interactions. The next decade will doubtless witness a flourish ofresearch papers on the impact of the Web as a communications medium, andmuch of our current wisdom about communicating science is likely to dra-matically change.

Journalists

Weaver and Wilhoit (1996) put the number of U.S. reporters and journal-ists at about 122,000, but only a small percentage of these reporters have sci-ence beats. There are only 600 to 800 individuals who are estimated to workas science and medical reporters (Klaidman 1991) and only about 2,000 indi-viduals who are members of the National Association of Science Writers,which includes print and broadcast journalists, freelance writers, and publicinformation officers (National Association of Science Writers 2001).

Few journalists covering science possess strong science backgrounds(Palen 1994); more commonly, writers learn science on the job (Hartz andChappell 1997). The vast majority of reporters do have college degrees (84 per-cent working for newspapers, 95 percent working for newsmagazines), butrarely do they have science degrees. Weaver and Wilhoit (1996) reported thatmore than 56 percent of journalists with college degrees majored in a com-munications-related field, while less than 3 percent majored in mathematics,physical science, or biological science. The situation is similar at the editoriallevel: fewer than one in three editors in a Canadian study had taken a singlescience course in college (Dubas and Martel 1975).


Encouragingly, specialized science reporters tend to be better educated inscience when compared to their general news peers. However, because theylack status in news organizations, it is likely to be a general reporter, not thescience writer, who is given the assignment when a fast-breaking news storydeals with science. Science reporters hold somewhat different news valuesthan regular reporters, favoring alternatives to hard news because such alter-native formats allow more effective communication about science issues(Friedman 1986; Glynn 1988).

Science writers approach their task differently depending on organiza-tional constraints. Dunwoody (1979) found that reporters covering theannual meeting of the American Association for the Advancement of Science(AAAS) who were working under strict deadlines were more dependent onpress conferences and therefore on the sponsoring organization than werereporters with fewer time constraints. In addition, the more stories a reporterwas expected to write, the greater was the reporter’s reliance on press confer-ences. The majority of stories produced by reporters with daily deadlineswere single-source stories, while the majority of stories written by reporterswith fewer constraints involved two or more sources. Since good reportingrequires input from several sources (Rubin and Hendy 1977), why do manyreporters rely on one or two? It may be because they often do not know whereto find sources for science-related issues (Friedman 1986).

Groups of prominent writers at science conferences may form an “innerclub,” with those writing for elite papers at the top (Dunwoody 1980). Thesewriters pool resources in deciding what to cover and how such coverage is tobe formulated. Altimore (1982) echoed this theme when he wrote that sci-ence reporting “is characterized by an inordinate degree of collaboration andcommunication among reporters, and science journalism is quite homoge-neous in its view of what qualifies as science news” (p. 25).

Science writers and their editors do not always agree on the types of sci-ence stories readers will find interesting (Dunwoody 1986b). And editors, ascompared to science writers, scientists, and lay persons, are more likely tofavor sensationalism and less likely to favor accuracy in judging newsworthi-ness (Johnson 1963). Dubas and Martel (1975) found that city editors tendnot to be very discriminating in selecting science stories, preferring storieswith a sensational angle or an element of conflict, or in some cases dismissingthe relevance of science stories altogether. Not surprisingly then, many sci-ence writers are unhappy with the priorities of their editors (Dennis andMcCartney 1979), believing they like to scare readers, ignore continuing sto-ries, and waste space and air time on junk. At the same time, since editorsoften write story headlines and control story revisions, science reporterssometimes write for editors rather than the public (Friedman 1986).


Science Information Professionals

Many times, the communications link from scientist to reporter will travelthrough a science information professional, or public relations person (alsoknown as public information, public relations, communications, publicaffairs, news service, and media relations). Science information profession-als are common in most large scientific societies, universities, major researchlaboratories, and industrial organizations (Rogers 1986).

Science information professionals often have trained as reporters, mean-ing they likely have little or no formal education in science. Increasingly,young persons trained in science journalism end up working as science infor-mation professionals instead of as journalists, perhaps reflecting job marketrealities (Rogers 1986). Science information professionals serve as spokes-people for their organizations, frequently appearing before communitygroups and media. They may also run speakers’ bureaus and coordinate spe-cial events; produce brochures, booklets, or reports; act as advisers to topofficials within organizations; and help individual scientists work moreeffectively with media.

The professional may be asked to help interpret implications of new devel-opments, suggest ways of dealing with media, and suggest the kinds of infor-mation that should be released to a public. He or she may be asked to producehow-to books for scientists dealing with media. Often, the science informa-tion professional is a liaison between scientists and reporters. The role ofboundary spanner is difficult because of the conflicting roles of scientists andjournalists, yet it can be an effective one. For example, more than half of thescientists in one study reported that mediation occurred in their interactionswith reporters and that it resulted in more accurate stories (Dunwoody andRyan 1983). In addition, the science information professional orients report-ers to ongoing research activities within the organization. Research suggeststhat most reporters covering the AAAS meeting use news conferences to helpdetermine what is important, and content analyses of print and other mediashow that science information professionals are the major sources of infor-mation from the meetings.

Unfortunately, these individuals are often low in the hierarchy of theirown institutions. Their efforts are supported with small budgets and fewresources. They typically receive no credit for the stories about science thatappear in the news. In the worst cases, the professional’s news release may becarried verbatim but with a reporter’s byline. Scientists see the professionalsas too close to the media, journalists see them as “flacks” for scientific orga-nizations, and both view them as representatives of organizational adminis-tration (Rogers 1986).


Scientists

Burkett (1986) estimated that 3 million persons were employed in theUnited States as scientists in the early 1980s, and more recent estimates sug-gest that number has held relatively constant (Commission on Professionalsin Science and Technology 2001). The number of individuals with at least abachelor’s degree in science or technology who are employed is closer to10 million (National Science Board 2000). The science workforce is sup-ported by a large amount of public and private spending: U.S. research anddevelopment expenditures are estimated to be $227 billion as of 1998,although research and development spending as a percentage of grossdomestic product has declined since the early 1990s (National Science Board2000).

With some exceptions, most working scientists have little responsibilityfor dealing directly with the public. An elite group of scientists, however,especially those who publish in journals monitored by the press, are oftensought for interviews by media reporters. Among the journals regularlyscanned by science journalists are Science (the weekly journal of the AAAS),Nature, the New England Journal of Medicine, and the Journal of the Ameri-can Medical Association. These journals frequently “speak not only of thetechnical matters of science but also of policy, politics, and conscience”(Burkett 1986, 8). Some famous scientists also are given relatively directaccess to the public by news organizations. These “visible scientists” includeNobel Prize winners, heads of prestigious institutions, and administrators ofscience-oriented agencies and labs (Goodell 1977).

There is a widespread perception that scientists are not effective commu-nicators, at least when the audience is the general public. Dr. Neal Lane (citedin Hartz and Chappell 1997), former head of the National Science Founda-tion, claimed:

With the exception of a few people . . . we don’t know how to communicatewith the public. We don’t understand our audience well enough—we have nottaken the time to put ourselves in the shoes of a neighbor, the brother-in-law,the person who handles our investments—to understand why it’s difficult forthem to hear us speak. We don’t know the language and we haven’t practiced itenough. (P. 38)

Most scientists appear ready to improve their skills, since more than 80 per-cent of scientists in a recent survey said they were willing to take a course tohelp them learn to communicate better with journalists. Roughly the sameamount, 81 percent, are at least somewhat willing to make the effort to com-municate with the public (Hartz and Chappell 1997).


The scientist who wants to communicate directly with a public aboutissues of science faces several important hurdles. Perhaps the most basic ofthese is language. As recently as 1920, the language used in a journal such asNature would be comprehensible to literate audiences and would not sounddramatically different from other forms of literature. But now, scientific lan-guage has “diverged from the mainstream of literary language and dividedinto a large number of small, winding tributaries” (Shortland and Gregory1991, 12). Hence, the scientist must be skilled at translating ideas from thetechnical language of his or her discipline into a currency accessible to layaudiences.

Some people, including a number of scientists (Eron 1986), argue that sci-entists have a basic responsibility to interact with the public. Yet, scientistsare often reluctant to engage in public dialogue. Fellow scientists may lookdown on colleagues who go public, believing that science is best sharedthrough peer-reviewed publications. Scientists may also believe that broad-cast media are trivial, that scientists should be humble and dedicated to theirwork, that scientists should have neither the time nor the inclination to blowtheir own trumpets, that the rewards of a media career can compromise a sci-entist’s integrity, that the public may commandeer a story and distort it, andfinally that the public may get excited about the wrong side of the story(Shortland and Gregory 1991).

Audiences

Large numbers of American adults appear to be scientifically “illiterate”(Maienschein and students 1999), leaving many to conclude there is a “prob-lem” in science communication (Dornan 1988, 1990; Durant and Evans1989; Durant, Evans, and Thomas 1992; Hartz and Chappell 1997;Trachtman 1981). Ziman (1992) proposed three ways to view the problem:the deficiency model, the rational choice model, and the context model. Thedeficiency model suggests that widespread ignorance about science is a prob-lem because scientists in democratic societies depend on public goodwill forfunding and support. If ignorance of science can be reduced, the public’s atti-tude toward science will be positive, resulting in ever-increasing levels ofeconomic support. Ignorant publics are vulnerable to the antiscience mes-sages of those who would cut science funding. Since most adults encounterscience information only from media coverage, ignorance is best reduced viaeffective communication about science. Effective communication wouldhelp adult nonscientists to become more literate about what scientists know.The model’s appeal is enhanced by findings that show widespread scientific


illiteracy in major democracies (Hartz and Chappell 1997) and by evidencethat attitudes toward science may be growing more negative (Yankelovich1982).

But, the deficiency model has important problems, including what some(Gregory and Miller 1998; Trench 1998) claim is its top-down, science-centered approach. And, there may be logical problems with asserting that abody of knowledge exists ready to be communicated to the uninformed sincescience is not “a well-bounded, coherent entity, capable of being more or less‘understood’ ” (Ziman 1992, 15). Scientists themselves have no clear andconsistent notion of what science covers and often disagree about what it tellsus about the world.

A second perspective is the rational choice model. It asks, “What do peo-ple need to know in order to be good citizens—even to survive—in a culturelargely shaped by science?” (Ziman 1992, 16). Without sufficient knowl-edge, people might not live their lives optimally, or they might even turnagainst science. But, dilemmas plague this approach too. For example, givenconflicts among scientists over findings and theory, whose advice should befollowed? What advice is necessary? Where should such advice be located?

Finally, the context model asks, “What do people want to know in theirparticular circumstances?” (Ziman 1992, 17-18). This model requires under-standing of the context of scientific knowledge and how different people putit to use. Lewenstein (1992), Logan (1999), and Ziman (1992) have arguedthat science communication scholarship could benefit from adopting thisthird perspective.

National Science Foundation surveys report that almost 90 percent of U.S.adults claim to be interested in news about science and technology. Below thesurface though, evidence suggests that the public can be divided into at leastthree segments according to level of interest in science (Miller 1986; Prewitt1982). Miller (1986) originally estimated that about 20 percent of Americanadults are attentive to science policy. This group tends to be younger, male,better educated, and more likely to have taken a college-level science coursewhen compared to the broader population. There is also evidence that thisgroup is shrinking, as recent surveys now suggest attentives number between10 and 14 percent of the population (National Science Board 2000).

Another 44 percent of the public can be characterized as “science-inter-ested” (National Science Board 2000). These individuals have a relativelyhigh interest in science and technology but lack functional understanding ofthe process or terminology of science. Compared to the science-attentivepublic, science interesteds are slightly older, somewhat less educated, andless likely to have had a college-level science course (Miller 1986). In linewith Miller’s (1986) findings, Palen (1994) reported that 56 percent of


Americans are regular viewers of television programs on science, technol-ogy, or nature, and 38 percent read science news in a newspaper weekly.

While many people profess interest in science, the unfortunate reality isthat two-thirds of even the attentive public cannot pass a “relatively minimaltest of scientific literacy” (Miller 1983, cited in Miller 1986, 66). Among thepublic as a whole, knowledge levels are even lower: fewer than half of therespondents to a recent national survey could correctly answer whetherhumans lived at the same time as dinosaurs, electrons are smaller than atoms,antibiotics kill viruses, lasers work by focusing sound waves, or it takes theearth one year to travel around the sun (National Science Board 2000). But,science knowledge is not unique in this regard; Americans appear prettyignorant in other areas too. Popular books assure us that Americans do notknow much about history (Davis 1999b), geography (Davis 1999a), comput-ers (Gookin 1999), mathematics (Paulos 1988, 1995), or almost any specialtyarea.

Such findings raise questions about the content of public understanding ofscience, leading some to question, What should the public understand?Should the public know about recent developments in science? Should itexhibit science literacy (i.e., basic understanding of accepted scientific factsand theories)? Should the public understand the methods of science? Shouldit possess insight into the implications of scientific findings? Is it importantthat the public understand all of these things or some combination of them?

The traditional view holds that all citizens ought to be scientifically liter-ate, as a means of ensuring their full participation in science policy formula-tion. Yankelovich (1982), a proponent of this view, argues that the generalpublic must be a target of science communication. But others, includingPrewitt (1982) and Miller (1986), believe that science messages are oftenwasted when disseminated to the general public. They suggest segmentingthe public according to where individuals exist in a science hierarchy. At thetop are decisionmakers in government and policy with specialized scienceinformation needs. These decisionmakers increasingly have to make com-parative judgments about science and technology matters, which require ahigh degree of scientific literacy to ensure that wise science policies aredeveloped and implemented. The attentive public also requires an under-standing of the process of scientific study and a “functional understanding ofthe major constructs used in scientific discourse [for example, molecule,gene, cell]” (Miller 1986, 61). The information needs of the interested publicare more difficult to address. Miller (1986) speculated that any approach tocommunicating with this group should be nontechnical, simple, and picto-rial. Finally, there is little consensus about the information needs or wants ofthe nonattentive public.


Almost 80 percent of the attentive public watch news shows regularly, androughly the same proportion of the interested public watch the news. About75 percent of attentives regularly read the paper, but they are dissatisfied withthe science coverage they find there, and just 9 percent rate the paper as agood source of science news. About half of the attentive public are regularreaders of one or more science magazines, including National Geographicand Psychology Today. But, fewer than 10 percent are readers of general sci-ence magazines such as Science, Discover, or Scientific American (Miller1986).

Beyond basic scientific facts, it is interesting to consider what peopleunderstand about the work of science and about the lives of scientists. Sci-ence is not a visible occupation, and people rarely observe scientists at work.LaFollette (1990) analyzed how popular magazines appearing between 1910and 1950 presented images of science. She found that the valence of imagesof science and scientists have waxed and waned through the years. Maga-zines generally linked science to national progress and economic health, andthe general trend over fifty years is an increase in articles about science.

The attitudes people hold toward science appear to be complex as well.Angell (1996) argued that the United States is in the midst of a groundswell ofantiscience feeling, pointing to renewed opposition to the teaching of evolu-tion in public schools as an example. Ironically, and perhaps not coinciden-tally, such sentiments come at a time when people are more dependent on sci-ence than ever. Yankelovich (1982) reported that the present image of scienceis somewhat less positive than it was earlier in the century. A bare majoritynow agrees that “technology will find a way to solve the problems of society,”and fewer people agree that “everything has a scientific explanation.” Morerecent surveys show conflicting public attitudes: two-thirds of respondentsagree that “science is the best source of reliable knowledge about the world,”but almost 40 percent of the public agree that “technology has become dan-gerous and unmanageable” (National Science Board 2000).

Audience attitudes may be influenced by the tone of coverage as well. Ananalysis of biotechnology accounts from 1970 through 1996 (Lewenstein,Allaman, and Parthasarathy 1998) found that coverage has, over time, beenconsistent and emphasized positive outcomes. Findings from several studiessuggest that no single generalization about tone may be appropriate for allmedia at all times. The tone of elite media coverage of the theory of evolutionchanged from doubting to supportive during the early part of the twentiethcentury (Caudill 1987). Bowes and Stamm (1972) found that the tone ofmedia coverage of a flood control project became more positive following thegrowth of public opposition to the project.


Whether reporting acknowledges or fails to present controversy in scienceis related to the tone of news accounts, since controversy can signal a negativetone. Cole’s (1975) content analysis of newspapers during the 1950s, 1960s,and 1970s found controversy more likely in articles from the 1970s than fromthe earlier times. Collins (1987) suggested that television typically ignoresscience controversy unless a story relates to another current issue. In Cana-dian papers, meanwhile, the overall tone of science articles is positive,according to one content analysis (Einsiedel 1992).

Agenda-setting research posits that the prominence of issues in newsmedia can affect the salience given to the issue among audiences (McCombsand Shaw 1972, 1993). Pilisuk and Acredolo (1988) surveyed three commu-nities, concluding that regular use of broadcast media is unrelated to concernabout technological risk. Conversely, Albert (1986) suggested that magazinecoverage of AIDS in the early 1980s contributed to a climate of blame forthose who have the disease. At the same time, Baker (1986) contended earlynews coverage of AIDS at an elite newspaper influenced perceptions of thedisease as a legitimate social issue.

Mazur and Conant (1978) found that people who have heard about a pro-posed nuclear waste site are more opposed to it than are people who have notheard about it. Mazur (1981a, 1981b) concluded that media coverage of a sci-entific controversy increases public opposition to the technology, even whensuch coverage is not negative. McLeod, Glynn, and Griffin (1987) found thatgreater media use is associated with higher ratings of the importance ofenergy. Placing an issue high on a public’s issue agenda can carry benefits.For example, one study found that there was an increase in the early detectionof colon cancer following the extensive media coverage of then PresidentReagan’s colon cancer surgery (Brown and Potosky 1990). And, events suchas Earth Day can spur coverage of environmental issues, even as the coverageemphasizes some environmental problems at the expense of others (Bowmanand Hanaford 1977).

There is a great deal of science reporting about risk, and this is one area inwhich public interest seems high. The reasons for this are obvious. Scientificdiscoveries can help people to avoid health threats (encouraging people to eatbetter and exercise more), detect threats (new technologies can help withearly diagnosis of disease or illness), or identify threats (link radon to soil orlink cell phones and smoking to cancer). There seems to be broad agreementthat a distinction can be made between the “objective reality” of risks, as evi-denced by statistical estimates from experts, and social perceptions of risk(Bradbury 1989; Golding 1992; Renn 1992). The divergence of the two maybe, in part, an issue of the extent and the way in which risk is covered by thepress (Burnham 1987; Viscusi 1992).


Papers dealing with risk issues cover a diverse set of phenomena. Promi-nent are coverage of the Chernobyl incident and other nuclear issues (Burkett1986; Nimmo and Combs 1985; Norstedt 1991; Peters 1992; Peters et al.1990; Rossow and Dunwoody 1991; Rothman and Lichter 1987; Stephensand Edison 1982), AIDS and HIV precautions (Dunwoody and Neuwirth1991; Singer, Rogers, and Glassman 1991; Stroman and Seltzer 1989; Witte1995), asbestos (Freimuth and Van Nevel 1981), earthquakes (Atwood 1998)and other natural disasters such as Mount St. Helens (Burkett 1986), the envi-ronment (Schoenfeld 1979), technology generally (Pilisuk and Acredolo1988), water safety (Griffin, Neuwirth, and Dunwoody 1995; Kahlor,Dunwoody, and Griffin 1998), and food safety (Juanillo and Scherer 1995),including pesticides, color additives, dioxin leaching into milk from contain-ers, and growth hormones in animals (Juanillo and Scherer 1995).

Whereas the literature on science communication often portrays thereader as relatively passive and uninvolved, audiences for information aboutrisk are often portrayed as active (Grunig 1974). For example, in 1989, therewere 250 organized boycotts of food products, up from 100 to 150 in a typicalyear (Juanillo and Scherer 1995). Consumer confidence about the safety offood dropped from 81 percent to 65 percent between January and June of1989 (Mueller 1990). Policies about nuclear energy, food irradiation,tobacco legislation, waste disposal, needle exchanges, disease prevention,and many other concerns are often more affected by the perception of riskthan by the quantified predictions of experts. Among other things, percep-tions of risk are affected not only by statistical probabilities but also by feel-ings of dread and by the extent to which the threat is either well understood orunknown (Slovic 1992).

Since society must tolerate a degree of risk, “classical risk communicationessentially translates as advocacy for determining which risks are accept-able” (Juanillo and Scherer 1995, 278). When risks are identified or labeledas concerns, stakeholders, including “experts, policy makers, interest groups,and the general public” (Juanillo and Scherer 1995, 279), become involved indebates about policies that are designed to increase safety. Media, althoughnot explicitly mentioned in the list, deserve a place as well because informa-tion from media influences many risk perceptions (Slovic 1992; Viscusi1992).

Conflicts among the Players

The science communication literature offers many perspectives on waysin which the interests, goals, values, and routines of scientists and science


journalists clash. These differing values may, in part, be responsible for mis-understandings and disagreements that can hinder relationships betweenjournalists and scientists.

Journalists’ norms appear frequently to contradict those of scientists.Journalists are attracted to stories that feature controversy and to new, evententative, results that carry exciting potential. Norms of fairness lead journal-ists to balance views of a topic rather than appeal to a single authority, even ifa disparity exists in the qualifications of the sources. Reporters face strict,inflexible deadlines. No matter how technical or abstract the issue, a journal-ist must write in prose that appeals to the broadest possible audience. In addi-tion, journalists write knowing that their copy will be judged, edited, andscreened by an editor, who may not be interested in science (Shortland andGregory 1991).

Journalists may view scientists as narrowly focused, obscure, andself-absorbed. Scientists are specialists, involved in the minutia of a specificproblem that may represent a small piece of a much larger puzzle. This canmake it difficult for them to state why their most recent discovery is a news-worthy event or even a significant development. Scientists offer predictionsthat are tentative and qualified, which may seem incompatible with fosteringexcitement in a story. But, bringing scientific and reporting values into line isnot simply an issue of making scientists less humble in their writing. In manycases, the importance of scientific work is not immediately obvious. Inalmost all cases, new discoveries are only an incremental part of a largerundertaking (Valenti 1999).

An important value of science is objectivity, not so much in the choice ofquestions or theories, but in requiring tests that permit theoretically incom-patible outcomes. For scientists, hypotheses must be falsifiable, and tests ofthe hypotheses must be replicable, so that others working in the discipline,including those with contrary theoretical views, may subject theories to rig-orous scrutiny. Conversely, journalism is a subjective enterprise. Indeed,some news organizations, such as the Washington Post, have abandoned theidea of objectivity for the somewhat different concept of “fairness” (seehttp://www.presswise.org.uk/Objectivity.htm). Long-term enterprise stories(health, government performance, and quality of life) are those that lead toPulitzer Prizes, yet these typically adopt a value-laden point of view.

Journalists have a great deal of confidence in scientists, more than they doin their own profession or in other major institutions. Journalists disagree(80 percent) that scientists who give interviews are publicity seekers andagree (80 percent) that scientists are at least somewhat accessible. Looking inthe mirror, few journalists agree that a professional code for journalistsensures high standards. A substantial majority agree that the “biggest


problem with science reporting is that it only tells a small part of the wholestory” (Hartz and Chappell 1997).

Reporters do offer some specific criticisms of scientists. Some feel thatscientists lack incentives to talk to small newspapers and that scientists andindustry researchers have vested interests (Crisp 1986; Kiernan 1998). Rus-sell (1986) argued that “for scientists, science communication with a layaudience is almost always a secondary issue. Of first importance, from a pro-fessional standpoint, is the business of science itself” (p. 83). She chargedthat scientists can be difficult to track down and reluctant to return calls. Ifreached, scientists “talk in the most technical language possible and are fear-ful of being misquoted” (p. 84). Not escaping criticism are scientists who docooperate with reporters but

who might be considered a bit too helpful in their efforts to utilize the press.Some researchers are interested in popularizing not only science but also theirown reputations. They even seek out writers with the help of their own publicrelations agents. (p. 85)

Explaining why some scientists may make themselves available, Russellbelieves that “the overly cooperative category also includes scientists with acause to push or a political point of view to promote” (p. 85).

Conflict sometimes emerges between scientists and journalists over own-ership of information about science. Breaking news about science is oftenintroduced at controlled events, such as scientific meetings or press confer-ences, or in journals. But, reporters may also find out about important storiesvia leaks from politicians, the actions or suspicious activities of key players,and articles in small trade publications. Conflict between scientists andreporters can emerge when scientists or journal editors attempt to control theinformation, for example, through the use of news embargoes (Kiernan1998).

In addition, scientists may hold that the emphasis on newsworthiness cancreate distortions in the reporting of scientific findings, characterized by thecharge of “sensationalistic” coverage (Gorney 1992). Scientists claim thatmedia coverage should educate and provide complete, nuanced descriptionsof scientific findings (Friedman, Dunwoody, and Rogers 1986). But, the sci-entist may find that:

the media have other agendas, and public education per se is not necessarilyprimary among them. Thus, efforts to inform the public about research inadvance are unlikely to succeed, because in the absence of controversy, scan-dal, or—yes—violence, it isn’t considered news. (West 1986, 40)


Scientists may also perceive that journalists ignore the balance of scien-tific evidence, giving equal weight to those presenting a broad scientific con-sensus with “maverick” scientists (Crisp 1986; Dearing 1995; Nelkin 1995).

An analysis of the poor quality of press coverage of scientific findingsconcerning violence and mass media concluded that researchers and report-ers have different responsibilities to different audiences, peers, and employ-ers (Eron 1986). The scientist’s primary responsibilities are to disseminateinformation, educate the public, be scientifically accurate, not lose facebefore colleagues, get some public credit for years of research, repay the tax-payers who supported the research, and break out of the ivory tower for thesheer fun of it. The journalist’s goals are to get the news, inform, entertain,not lose face before his or her colleagues, fill space or time, and not be repeti-tive. Sometimes these divergent agendas work to mutual benefit, but at othertimes they lead to conflict (Tavris 1986).

A recent survey of scientists and journalists confirmed that scientists holdnegative views of reporters (Hartz and Chappell 1997). For example, only11 percent of scientists have a great deal of confidence in the press, while22 percent have hardly any confidence. More than nine out of ten scientistsagree that few reporters understand the nature of science and technology,especially the “tentativeness of most scientific discovery and the complexi-ties of the results” (p. 29).

Scientists view themselves in a far more positive light. Almost 77 percenthave a great deal of confidence in themselves and their colleagues, while80 percent disagree that they waste the taxpayers’ money. Most (72 percent)want the public to know about their work, but a significant minority (40 per-cent) is afraid of being embarrassed before their peers by news stories abouttheir work. Most are willing to talk with reporters, but hardly any actually doso on a regular basis (only 4 percent as often as once a month). More than aquarter of the scientists from the sample have never appeared in the popularpress.

Differences in defining the boundaries of legitimate science also cancause conflict. Griffin and Dunwoody (1995) examined how advocacygroups provide information subsidies to news organizations in an effort to getcoverage of an issue the group believes important. Their work raises a moregeneral issue frequently ignored in the science communication literature,namely the influence of nonscientists on ways that publics encounter sciencenews. In fact, journalists frequently adopt (in the scientists’ view) an overlybroad definition of who is qualified to comment on scientific issues. Thus,political activists for issues such as animal rights, nuclear power, the environ-ment, the educational system, disease-afflicted groups, and so on may be pre-sented to the public as qualified experts on issues of science. This raises a


problem of evidence versus assertions. Reporters rarely ask how sourcesknow what they know, or what evidence the knowledge is based on, or why itdiffers from conventional wisdom (Tavris 1986):

Knowing little about methods and the differences among psychological disci-plines . . . many media people have never learned to be critical, what questionsto ask. Moreover, it is the nature of the social science to produce many contra-dictory, conflicting studies. To journalists, however, it often seems as though ifthey don’t like what one report says, another study will confirm their prejudicesand appear in 20 minutes. (Tavris 1986, 24-25)

Dunwoody (1986a) explored the issue of the costs and benefits for a scien-tist wishing to use the mass media to communicate science. There is great riskfor scientists because while they will find few tangible rewards for informingthe public, there are many concrete costs. Within the scientific community,public communication activities are seen as distracting from efforts to doresearch or even as grandstanding. In addition, public understanding of sci-ence carries little currency among scientists.

Relatively few empirical studies have examined direct contact of mediaand scientific organizations. In one, DiBella, Ferri, and Padderud (1991) sug-gested the primary motive of scientists for giving media interviews is to helpeducate the public about science. In another, Dunwoody and Ryan (1987)found that while scientists are generally asked by the press to comment ontopics related to their research expertise, about one-third of science-reporterinteractions deal with issues having little or no relationship to the scientist’sresearch.

Scientists fear that their own culture does not value direct contact with thepublic via general news media (Dunwoody and Ryan 1985). Although thismight suggest an important role for intermediaries such as public informationprofessionals, Dunwoody and Ryan (1985) found that scientists, whileexpressing a positive attitude toward public information professionals, con-sider them to be of only modest importance in disseminating informationabout science.

How Can Science Reporting Be Improved?

Improving science communication may involve changes in the way thatscience journalism is practiced. Indeed, the very label “science journalism”obscures the diverse activities ranging from coverage of basic science in spe-cialty magazines to reports on important science stories at elite papers andlocal news accounts of emerging local issues with a technological angle.


Journalist Training

Journalists almost always lack science training. One study suggests that“journalists tend not to have even a liberal-arts background in the sciences.Few understand the scientific method, the dictates of peer review, the reasonsfor the caveats and linguistic precision scientists employ when speaking oftheir work” (Hartz and Chappell 1997, 22). When a journalist lacks the back-ground to evaluate or understand complicated scientific issues, he or she isforced to deal with the subset of available scientists who are skillful at trans-lating complicated issues into simple prose. But, such sources may be quiterare. An alternative is for news organizations to invest in or at least to expectbetter training in science and technology from their reporters. It is common,for example, for university journalism programs to require that studentsdevelop a basic familiarity with the workings of government, with communi-cation law and policy, and with history. Far less common is a requirement forbasic scientific or mathematical literacy.

Although it may seem obvious that improving the science training ofreporters will enhance the quality of science journalism, this proposal is con-troversial (Hartz and Chappell 1997). Those who resist the idea offer the fol-lowing rationales: (1) some outstanding science reporting is done by individ-uals with little formal training in science, (2) reporters with excellentreporting skills can get scientific sources to explain research in simple andaccessible terms, and (3) it is impractical for most people to receive enoughtraining to serve as an expert across multiple disciplines of science, such aschemistry, biology, medicine, psychology, engineering, and physics. One canaccept the first two points and still believe that science reporting is improvedwhen journalists receive training in a scientific discipline. The final pointintroduces a more difficult issue, and the task of covering all of science maybe too broad for one person. Rather, such coverage may require specializa-tion, the same way reporters may specialize in covering the White House,Capitol Hill, and the Supreme Court.

Finally, recent work (Trumbo, Dunwoody, and Griffin 1998) hasapproached the issue of inaccurate science reporting from a different per-spective. This perspective locates accuracy problems as originating withfairly well-understood cognitive limitations on the part of reporters. This lineof research may be important in shifting the debate away from norm differ-ences (which are unlikely to change) and toward better reporter training as away of improving science coverage. Specifically, reporters could be trainedto identify and overcome the cognitive shortcuts they use that lead to inaccu-rate news accounts.


Focus on Audience Needs

Crane (1992) advocated that reporters take advantage of preexisting inter-ests of audiences. She suggested that reporters routinely specify for the audi-ence, “Why is it important for me to know about this story?” (p. 29). Sciencejournalism should also characteristically provide more background informa-tion and provide perspectives on what a story implies for the broader society.Effective science journalism should provide new information and connectscience to everyday life (Bostian 1983; Bostian and Byrne 1984; Hunsaker1979).

Another way to improve communication is to focus on style. Conventionsof journalistic style date to formulas generated in the 1930s: simple words,short sentences, and an inverted pyramid for organizing information. Yet,according to Dunwoody (1992), “to this day in the world of journalism, thereis very little attention paid to what people actually get out of the stuff that theyread. There are still a lot of assumptions being made” (p. 102). Rowan (1989,1991a, 1991b, 1992) presented a number of recommendations for improvingscience writing. She began by asking the journalist to focus on explanation,which she defined as “anticipating and overcoming likely confusions”(Rowan 1992, 131). Those who effectively explain possess a conviction inthe value of good explanation, large and easily accessed collections of expla-nations and conceptual frameworks for determining why ideas are likely tobe difficult for audiences, and strategies that best overcome these obstacles.

Interestingly, although accuracy in science reporting is an enduring con-cern (Ryan 1979; Ryan and Owen 1977; Singer 1990; Tankard and Ryan1974), the studies that have examined the issue often have found general sat-isfaction among scientists with news story accuracy. For example, newspapercoverage of research appearing in the New England Journal of Medicine wasgenerally accurate in distinguishing facts from opinion (Caudill andAshtown 1989). And, scientists asked to comment on science content in mag-azines (Borman 1978) and on television (Moore and Singletary 1985)reported accuracy was, on balance, good (Pulford 1976).

Work More Closely with Sources

Broberg (1973) examined the changes that scientists would make to sto-ries about their research and found that additions are the most common cor-rection. This is consistent with other research suggesting scientists’ majorcontention with press coverage concerns omissions rather than misstate-ments (Borman 1978; Dunwoody 1982). Greenberg and Wartenberg (1990)analyzed network news coverage of disease and teen suicide to determine


whether such coverage provided an accurate portrayal of the geography ofthe diseases and concluded the coverage was accurate (see also Greenberget al. 1989).

The task for journalists is daunting, however. Siegfried (1992) summed upthe constraints on newspaper coverage by writing, “The truth is that in dailynewspaper journalism there is very little room or place for any real explana-tion. Cancer is cured, fewer people will die—that’s the end of the story in thedaily newspaper” (p. 113). Ward (1992) argued the situation is even worse whenconsidering how science is covered on television news shows. The length ofthe average sound bite on television is six seconds. The importance of com-pelling visuals leads producers to require personalization (show me someonewho has got the disease) and a news peg.

Future Directions in Science Communication

Science communicators have mapped out an ambitious agenda. Sciencecommunication research has long been concerned with what people do withthe knowledge they gain from media. This strong record of achievement ismost clear in the area of risk communication, where investigators try to dis-cover how people react to information about technological threats. Moreresearch might be devoted to how people use other kinds of science knowl-edge. For example, how do people use information about astronomy, earthscience, physics, chemistry, and other topics that do not necessarily ordirectly involve risk? And, how should science activities with no immediatepayoff be framed and covered?

Even learning, as defined above, is too restrictive for representing sciencecommunication scholarship. A broad set of attitudinal questions is also pres-ent in the literature. These questions concern how people form attitudestoward science, scientists, technology and specific technologies, funding ofscience, science education, and science policy. Attitudinal and opinion issuessuch as these find their intellectual roots in persuasion theory, in theories ofpublic opinion, and in political science. Especially important and usefulwould be efforts to link specific science and technology attitudes to the typesof knowledge that people have about science.

A prevalent assumption in the literature is that high levels of knowledgecorrespond to favorable attitudes toward science (Schibeci 1990). But, thereare few data on this critical issue (Althoff, Grieg, and Stuckey 1973). It maybe just as logical to assume that moderately high levels of knowledge areassociated with antiscience attitudes. It does not seem unreasonable tohypothesize that radical environmentalists, antinuclear activists, opponents


of cloning and genetic research, animal rights protesters, and others whoexpress narrow or broad opposition to scientific research efforts might actu-ally possess greater levels of understanding of science and the scientificmethod than do members of the nonattentive public. This is not to say thatthese groups possess high levels of understanding, merely that they have atleast some understanding, if for no other reason than because they are fre-quently forced to defend their beliefs. This prediction suggests there may be acurvilinear (U-shaped) relation between science knowledge and science atti-tudes. If science communication is concerned with the kinds of knowledgethat foster greater appreciation of science, it also must be concerned with thekinds of knowledge that foster antiscience sentiments.

Attitudinally related questions should also be more specifically framed.One’s general attitude toward science (i.e., science is good) may be very dif-ferent from attitudes toward specific issues (cloning, space exploration), sci-entists (odd characters, role models), general science support (we are spend-ing enough on science now), and specific support (we need to spend less onAIDS research and more on cancer, or we should fund a greater number ofmodest physics experiments and fewer big experiments, or too much moneyis spent on medical research and not enough on chemical research). Under-standing people’s beliefs and attitudes about science would give us a muchbetter understanding of science publics. Attention should be given to theimplicit notion described by Ziman’s (1992) deficiency model, namely thatknowledge → attitude → funding. A failure to find such predicted linkswould confirm that issues of science literacy and support are quite differentand should be treated as distinct problems.

Science communication research has examined many sociological andpublic-policy questions. These include the sociology of news and factorsaffecting the behavior of reporters, sources, news organizations, scientists,and news publics. Increasingly, scholarship in this area is examining theimpact that nonscientists have on science-related questions. This is anextremely important area of research because it is centrally related to sciencepolicy. What role do activists play in science communication? How dodecisionmakers obtain their science news? How is science policy made?When does the public play a role in science policy? What issues dopolicymakers contend with in decisions to support science and specificresearch activities? How do journalists balance their needs for close workingrelationships with scientists with their needs for autonomy? Do journalisticnorms for the coverage of government and policymakers hinder or enhancethe quality of coverage of scientists? What is the effect of a reporter’s ownscience literacy on his or her coverage of science, selection of stories, choiceof sources, and quality of reporting?


Science communication has immeasurably enhanced our knowledge notonly about how science information is communicated but also about masscommunication processes more generally. This healthy and vibrant area ofscholarship is likely to become even more central to the discipline of masscommunication. The special challenges presented by communicating thecomplex and important issues of science will only grow in importance in thisnew century.

References

Albert, E. 1986. Acquired immune deficiency syndrome: The victim and the press. In Studies incommunications, Vol. 3, edited by T. McCormack, 135-58. Greenwich, CT: JAI.

Altheide, D. L. 1976. Creating reality: How TV news distorts events. Beverly Hills, CA: Sage.Althoff, P., W. H. Grieg, and F. Stuckey. 1973. Environmental pollution control attitudes of

media managers in Kansas. Journalism Quarterly 50:666-72.Altimore, M. 1982. The social construction of a scientific controversy: Comments on press cov-

erage of the recombinant DNA debate. Science, Technology, & Human Values 7:24-31.Angell, M. 1996. Science on trial: The clash of medical evidence. New York: Norton.Atwood, L. E. 1998. Risk perception and information management responses to a predicted

earthquake. Paper presented at the annual meeting of the Association for Education in Jour-nalism and Mass Communication, Baltimore, August.

Bader, R. G. 1990. How science news sections influence newspaper science coverage: A casestudy. Journalism Quarterly 67:88-96.

Baker, A. J. 1986. The portrayal of AIDS in the media: An analysis of articles in the New YorkTimes. In The social dimension of AIDS: Method and theory, edited by D. A. Feldman andT. M. Johnson, 179-94. New York: Praeger.

Banks, J., and J. D. Tankel. 1990. Science as fiction: Technology in prime time television. Criti-cal Studies in Mass Communication 7:24-36.

Boorstin, D. J. 1983. The discoverers: A history of man’s search to know his world and himself.New York: Random House.

Borman, S. C. 1978. Communication accuracy in magazine science reporting. Journalism Quar-terly 55:345-46.

Bostian, L. R. 1983. How active, passive and nominal styles affect readability of science writing.Journalism Quarterly 60:635-40, 670.

Bostian, L. R., and T. E. Byrne. 1984. Comprehension of styles of science writing. JournalismQuarterly 61:676-78.

Bowes, J. E., and K. R. Stamm. 1972. Communicating during an environmental decision. Jour-nal of Environmental Education 3:49-55.

Bowman, J. S., and K. Hanaford. 1977. Mass media and the environment since Earth Day. Jour-nalism Quarterly 54:160-65.

Bradbury, J. 1989. The policy implications of differing concepts of risk. Science, Technology, &Human Values 14:380-99.

Breed, W. 1980. The newspaperman, news, and society. New York: Arno.Broberg, K. 1973. Scientists’ stopping behavior as indicator of writer’s skill. Journalism Quar-

terly 50:763-67.


Brown, M. L., and A. L. Potosky. 1990. The presidential effect: The public health response tomedia coverage about Ronald Reagan’s colon cancer episode. Public Opinion Quarterly50:317-29.

Burkett, W. 1986. News reporting: Science, medicine, and high technology. Ames: Iowa StateUniversity Press.

Burnham, J. 1987. How superstition won and science lost: Popularizing science and health inthe United States. New Brunswick, NJ: Rutgers University Press.

Caudill, E. 1987. A content analysis of press views of Darwin’s evolution theory. JournalismQuarterly 64:782-86.

. 1989. The roots of bias: An empiricist press and coverage of the Scopes trial. Journal-ism Monographs 114.

Caudill, E., and P. Ashtown. 1989. The New England Journal of Medicine as news source. Jour-nalism Quarterly 66:458-62.

Cole, B. J. 1975. Trends in science and conflict coverage in four metropolitan newspapers. Jour-nalism Quarterly 52:465-71.

Collins, H. M. 1987. Certainty and the public understanding of science. Social Studies of Science17:689-713.

Commission on Professionals in Science and Technology. 2001. Scientists and engineers for thenew millennium: Renewing the human resource. Retrieved 30 July 2001: http://www.cpst.org/S&E.PDF.

Crane, V. 1992. Listening to the audience: Producer-audience communication. In When sciencemeets the public, edited by B. V. Lewenstein, 21-32. Washington, DC: American Associa-tion for the Advancement of Science.

Crisp, D. W. 1986. Scientists and the local press. In Scientists and journalists: Reporting scienceas news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers, 73-80. New York: FreePress.

Cronholm, M., and R. Sandell. 1981. Scientific information: A review of research. Journal ofCommunication 31:85-96.

David, P. 1996. Role of imagery in recall of deviant news. Journalism and Mass CommunicationQuarterly 73:804-19.

Davis, K. G. 1999a. Don’t know much about geography. New York: Avon.. 1999b. Don’t know much about history. New York: Avon.Dearing, J. W. 1995. Newspaper coverage of maverick science: Creating controversy through

balancing. Public Understanding of Science 4:341-61.Dennis, E., and J. McCartney. 1979. Science journalists on metropolitan dailies. Journal of Envi-

ronmental Education 10:10-11.DiBella, S. M., A. J. Ferri, and A. B. Padderud. 1991. Scientists’ reasons for consenting to mass

media interviews: A national survey. Journalism and Mass Communication Quarterly68:740-49.

Dornan, C. 1988. The “problem” of science and media: A few seminal texts in their context,1956-1965. Journal of Communication Inquiry 12:53-70.

. 1990. Some problems in conceptualizing the issue of “science and the media.” CriticalStudies in Mass Communication 7:48-71.

Dubas, O., and L. Martel. 1975. Media impact: A research study on science communication.Ottawa, Canada: Ministry of State.

Dunwoody, S. 1979. News-gathering behaviors of specialty reporters: A two-level comparisonof mass media decision-making. Newspaper Research Journal 1:29-41.

. 1980. The science writing inner club: A communication link between science and thelay public. Science, Technology, & Human Values 5:14-22.


. 1982. A question of accuracy. IEEE Transactions on Professional CommunicationPC-25:196-99.

. 1986a. The scientist as source. In Scientists and journalists: Reporting science as news,edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers, 3-16. New York: Free Press.

. 1986b. When science writers cover the social sciences. In Reporting science: The caseof aggression, edited by J. H. Goldstein, 67-81. Hillsdale, NJ: Lawrence Erlbaum.

. 1992. Comparative strategies for making the complex clear. In When science meets thepublic, edited by B. V. Lewenstein, 101-2. Washington, DC: American Association for theAdvancement of Science.

Dunwoody, S., and K. Neuwirth. 1991. Coming to terms with the impact of communication ofscientific and technological risk judgments. In Risky business: Communicating issues of sci-ence, risk, and public policy, edited by L. Wilkins and P. Patterson, 11-30. Westport, CT:Greenwood.

Dunwoody, S., and M. Ryan. 1983. Public information persons as mediators between scientistsand journalists. Journalism Quarterly 60:647-56.

. 1985. Scientific barriers to the popularization of science in the mass media. Journal ofCommunication 35 (1): 26-42.

. 1987. The credible scientific source. Journalism Quarterly 64:21-27.Durant, J., and G. A. Evans. 1989. Understanding of science in Britain and the U.S.A. In British

social attitudes: Special international report, edited by R. Jowell, S. Witherspoon, andL. Brook. Aldershot, UK: Gower.

Durant, J., G. A. Evans, and G. P. Thomas. 1992. Public understanding of science in Britain: Therole of medicine in the popular representation of science. Public Understanding of Science1:161-82.

Einsiedel, E. F. 1992. Framing science and technology in the Canadian Press. Public Under-standing of Science 1:89-101.

Entwistle, V., and M. Hancock-Beaulieu. 1992. Health and medical coverage in the UK NationalPress. Public Understanding of Science 1:164-88.

Eron, L. D. 1986. The social responsibility of the scientist. In Reporting science: The case ofaggression, edited by J. H. Goldstein. Hillsdale, NJ: Lawrence Erlbaum.

Evans, W. A., M. Krippendorf, J. H. Yoon, P. Posluszny, and S. Thomas. 1990. Science in theprestige and national tabloid presses. Social Science Quarterly 71:105-17.

Ferris, T. 1997. The whole shebang: A state-of-the-universes report. New York: Touchstone.Freimuth, V. S., and J. P. Van Nevel. 1981. Reaching the public: The asbestos awareness cam-

paign. Journal of Communication 31:155-67.Friedman, S. M. 1986. The journalist’s world. In Scientists and journalists: Reporting science as

news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers, 17-41. New York: FreePress.

Friedman, S. M., S. Dunwoody, and C. L. Rogers. 1986. Introduction. In Scientists and journal-ists: Reporting science as news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers,xi-xvii. New York: Free Press.

Gans, H. 1979. Deciding what’s news. New York: Random House.Gerbner, G. 1987. Science on television: How it affects public conceptions. Issues in Science and

Technology 3:109-15.Gleick, J. 1987. Chaos: Making a new science. New York: Viking.Glynn, C. J. 1988. Science reporters and their editors judge “sensationalism.” Newspaper

Research Journal 6:69-74.Golding, D. 1992. A social and programmatic history of risk research. In Social theories of risk,

edited by S. Krimsky and D. Golding, 23-52. Westport, CT: Praeger.


Goodell, R. 1977. The visible scientists. Boston: Little, Brown.Gookin, D. 1999. PCs for dummies. New York: Hungry Minds.Gorney, C. M. 1992. Numbers versus pictures: Did network television sensationalize Chernobyl

coverage? Journalism Quarterly 69:455-65.Gould, S. J. 1995. Dinosaur in a haystack: Reflections in natural history. New York: Crown.Greenberg, M. R., D. B. Sachsman, P. M. Sandman, and K. L. Salomone. 1989. Risk, drama and

geography in coverage of environmental risk by network TV. Journalism Quarterly 66:267-76.

Greenberg, M. R., and D. Wartenberg. 1990. Network television evening news coverage ofinfectious disease events. Journalism Quarterly 67:142-46.

Gregory, J., and S. Miller. 1998. Science in public: Communication, culture, and credibility.New York: Plenum.

Griffin, R. J., and S. Dunwoody. 1995. Impacts of information subsidies and community struc-ture on local press coverage of environmental contamination. Journalism and Mass Commu-nication Quarterly 72:271-84.

Griffin, R. J., K. Neuwirth, and S. Dunwoody. 1995. Using the theory of reasoned action toexamine the impact of health risk messages. In Communication Yearbook 18, edited by B. R.Burleson, 201-28. Thousand Oaks, CA: Sage.

Grube, A., and K. Boehme-Duerr. 1988. AIDS in international news magazines. JournalismQuarterly 65:686-89.

Grunig, J. E. 1974. Three stopping experiments on the communication of science. JournalismQuarterly 51:387-99.

. 1979. Research on science communication: What is known and what needs to beknown. ACE Quarterly 62:17-45.

. 1983. Communication behaviors and attitudes of environmental publics: Two studies.Journalism Monographs 81.

Hartz, J., and R. Chappell. 1997. Worlds apart: How the distance between science and journal-ism threatens America’s future. Nashville, TN: First Amendment Center.

Hawking, S. W. 1988. A brief history of time: From the big bang to black holes. New York:Bantam.

Hilgartner, S., and C. L. Bosk. 1988. The rise and fall of social problems: A public arenas model.American Journal of Sociology 94:53-78.

Hunsaker, A. 1979. Enjoyment and information gain in science articles. Journalism Quarterly56:617-19.

Jerome, F. 1986. Prospects for science journalism. In Reporting science: The case of aggression,edited by J. H. Goldstein, 147-54. Hillsdale, NJ: Lawrence Erlbaum.

Johnson, K. G. 1963. Dimensions of judgment of science news stories. Journalism Quarterly40:315-22.

Juanillo, N. K., and C. W. Scherer. 1995. Attaining a state of informed judgments: Toward a dia-lectical discourse on risk. In Communication Yearbook 18, edited by B. R. Burleson, 278-99.Thousand Oaks, CA: Sage.

Kahlor, L., S. Dunwoody, and R. J. Griffin. 1998. The role of attribution and framing in rational-izing about risk estimates. Paper presented at the annual meeting of the Association for Edu-cation in Journalism and Mass Communication, Baltimore, August.

Kiernan, V. 1998. Mars meteorite: A case study in controls on dissemination of science news.Paper presented at the annual meeting of the Association for Education in Journalism andMass Communication, Baltimore, August.

Klaidman, S. 1991. Health in the headlines: The stories behind the stories. New York: OxfordUniversity Press.


LaFollette, M. C. 1990. Making science our own: Public images of science 1910-1955. Chicago:University of Chicago Press.

Lewenstein, B. V. 1992. Introduction. In When science meets the public, edited by B. V.Lewenstein, ix-xvi. Washington, DC: American Association for the Advancement ofScience.

Lewenstein, B. V., T. Allaman, and S. Parthasarathy. 1998. Historical survey of media coverageof biotechnology in the United States 1970 to 1996. Paper presented at the annual meeting ofthe Association for Education in Journalism and Mass Communication, Baltimore, August.

Liebler, C. M., and J. Bendix. 1997. How distant the forest? Proximity, environmental contro-versy and source status conferral. Paper presented at the annual meeting of the Associationfor Education in Journalism and Mass Communication, Chicago, August.

Logan, R. 1999. Review essay: Worlds apart. Science Communication 20:337-43.Maienschein, J., and students. 1999. Commentary: To the future—Arguments for scientific liter-

acy. Science Communication 21:75-87.Mazur, A. 1981a. The dynamics of technical controversy. Washington, DC: Communications

Press.. 1981b. Media coverage and public opinion on scientific controversies. Journal of Com-

munication 31:106-15.Mazur, A., and B. Conant. 1978. Controversy over a local nuclear waste repository. Social

Studies of Science 8:235-43.McCombs, M. E., and D. S. Shaw. 1972. The agenda-setting function of mass media. Public

Opinion Quarterly 36:176-87.. 1993. The evolution of agenda-setting research: Twenty-five years in the marketplace of

ideas. Journal of Communication 43:58-67.McLeod, J. M., C. J. Glynn, and R. J. Griffin. 1987. Communication and energy conservation.

Journal of Environmental Education 18:29-37.Miller, J. D. 1983. Scientific literacy: A conceptual and empirical review. Daedalus 112:29-48.. 1986. Reaching the attentive and interested publics for science. In Scientists and jour-

nalists: Reporting science as news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rog-ers, 55-69. New York: Free Press.

Moore, B., and M. Singletary. 1985. Scientific sources’ perceptions of network news accuracy.Journalism Quarterly 62:816-23.

Mueller, W. W. 1990. Who’s afraid of food? American Demographics 12:40-43.National Association of Science Writers. 2001. NASW brochure. Retrieved 30 July 2001:

http://www.nasw.org/brochure.htm.National Geographic Society. 1976. Our continent: A natural history of North America. Wash-

ington, DC: National Geographic Society.National Science Board. 2000. Science and engineering indicators 2000. Available: http://www.

nsf.gov/sbe/srs/seind00/.Nelkin, D. 1995. Selling science: How the press covers science and technology. Rev. ed. New

York: Freeman.Nimmo, D., and J. E. Combs. 1985. Nightly horrors. Knoxville: University of Tennessee Press.Norstedt, S. A. 1991. The information crisis in Sweden after Chernobyl. Media, Culture, and

Society 13:477-97.Palen, J. A. 1994. A map for science reporters: Science, technology, and society studies concepts

in basic reporting and newswriting textbooks. Michigan Academician 26:507-19.Paulos, J. A. 1988. Innumeracy: Mathematical illiteracy and its consequences. New York:

Vintage.Paulos, J.P. 1995. A mathematician reads the newspaper. New York: Basic Books.


Peters, H. P. 1992. The credibility of information sources in West Germany after the Chernobyldisaster. Public Understanding of Science 1:325-43.

Peters, H. P., G. Albrecht, L. Hennen, and H. U. Stegelmann. 1990. “Chernobyl” and the nuclearpower issue in West German public opinion. Journal of Environmental Psychology10:121-34.

Pilisuk, M., and C. Acredolo. 1988. Fear of technological hazards: One or many? Social Behav-ior 3:17-24.

Pinker, S. 1994. The language instinct: How the mind creates language. New York: WilliamMorrow.

Prewitt, K. 1982. The public and science policy. Science, Technology, & Human Values 36:5-14.Pulford, D. L. 1976. A follow-up study of science news accuracy. Journalism Quarterly

53:119-21.Ramsey, S. A. 1989. The role of technological development in setting the stage for expanded

local science coverage. Southwestern Mass Communication Journal 5:33-40.Renn, O. 1992. Concepts of risk: A classification. In Social theories of risk, edited by S. Krimsky

and D. Golding, 53-82. Westport, CT: Praeger.Rogers, C. L. 1986. The practitioner in the middle. In Scientists and journalists: Reporting sci-

ence as news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers, 42-54. New York:Free Press.

Rossow, M. D., and S. Dunwoody. 1991. Inclusion of “useful” detail in newspaper coverage of ahigh-level nuclear waste siting controversy. Journalism Quarterly 68:87-100.

Rothman, S., and S. R. Lichter. 1987. Elite ideology and risk perception in nuclear energy policy.American Political Science Review 81:383-404.

Rowan, K. E. 1989. Moving beyond the what to why: Differences in professional and popularscience writing. Journal of Technical Writing and Communication 19:161-79.

. 1991a. Goals, obstacles and strategies in risk communication: A problem-solvingapproach to improving communication about risk. Journal of Applied CommunicationResearch 19:300-329.

. 1991b. When simple language fails: Presenting difficult science to the public. Journalof Technical Writing and Communication 21:369-82.

. 1992. Strategies for enhancing comprehension of science. In When science meets thepublic, edited by B. V. Lewenstein, 131-43. Washington, DC: American Association for theAdvancement of Science.

Rubin, D. M., and V. Hendy. 1977. Swine influenza and the news media. Annals of Internal Med-icine 87:772-73.

Russell, C. 1986. The view from the national beat. In Scientists and journalists: Reporting sci-ence as news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rogers, 81-94. New York:Free Press.

Ryan, M. 1979. Attitudes of scientists and journalists towards media coverage of science news.Journalism Quarterly 56:18-26.

Ryan, M., and D. Owen. 1977. An accuracy survey of metropolitan newspaper coverage of socialissues. Journalism Quarterly 54:27-32.

Sagan, C. 1980. Cosmos. New York: Random House.Schibeci, R. A. 1990. Public knowledge and perceptions of science and technology. Bulletin of

Science, Technology, and Society 10:86-92.Schoenfeld, A. C. 1979. The press and NEPA: The case of the missing agenda. Journalism Quar-

terly 56:577-85.Shoemaker, P. J., and S. D. Reese. 1991. Mediating the message: Theories of influences on mass

media content. White Plains, NY: Longman.


Shortland, M., and J. Gregory. 1991. Communicating science: A handbook. New York: Longman.Siegfried, T. 1992. Newspapers. In When science meets the public, edited by B. V. Lewenstein,

113-24. Washington, DC: American Association for the Advancement of Science.Singer, E. 1990. A question of accuracy: How journalists and scientists report research on haz-

ards. Journal of Communication 40:102-16.Singer, E., P. Endreny, and M. B. Glassman. 1991. Media coverage of disasters: Effects of geo-

graphic location. Journalism Quarterly 68:48-58.Singer, E., T. F. Rogers, and M. B. Glassman. 1991. Public opinion about AIDS before and after

the 1988 U.S. government public information campaign. Public Opinion Quarterly55:161-79.

Slovic, P. 1992. Perception of risk: Reflections on the psychometric paradigm. In Social theoriesof risk, edited by S. Krimsky and D. Golding, 117-52. Westport, CT: Praeger.

Stephens, M., and N. G. Edison. 1982. News media coverage of issues during the accident atThree-Mile Island. Journalism Quarterly 59:199-204.

Stroman, C. A., and R. Seltzer. 1989. Mass media use and knowledge of AIDS. JournalismQuarterly 66:881-87.

Tankard, J. W., and M. Ryan. 1974. News source perceptions of accuracy of science coverage.Journalism Quarterly 51:219-25, 334.

Tavris, C. 1986. How to publicize science: A case study. In Reporting science: The case ofaggression, edited by J. H. Goldstein, 21-32. Hillsdale, NJ: Lawrence Erlbaum.

Trachtman, L. 1981. The public understanding of science effort: A critique. Science, Technol-ogy, & Human Values 36:10-15.

Trench, B. 1998. Science reporting in Europe: From comparison to critique. Paper presented atthePublicCommunicationofScienceandTechnologyConference,Berlin,Germany,September.

Trumbo, C. W., S. Dunwoody, and R. J. Griffin. 1998. Journalists, cognition, and the presenta-tion of an epidemiologic study. Science Communication 19:238-65.

Valenti, J. 1999. Commentary: How well do scientists communicate to media? Science Commu-nication 21:172-78.

Viscusi, K. 1992. Smoking: Making the risky decision. New York: Oxford.Ward, J. 1992. Television. In When science meets the public, edited by B. V. Lewenstein,

103-106. Washington, DC: American Association for the Advancement of Science.Weaver, D. H., and G. C. Wilhoit. 1996. The American journalist in the 1990s: U.S. news people

at the end of an era. Mahwah, NJ: Lawrence Erlbaum.West, L. J. 1986. How not to publicize research: The UCLA Violence Center. In Reporting science:

The case of aggression, edited by J. H. Goldstein, 67-81. Hillsdale, NJ: Lawrence Erlbaum.Witte, K. 1995. Generating effective risk messages: How scary should your risk communication

be? In Communication Yearbook 18, edited by B. R. Burleson, 229-54. Thousand Oaks, CA:Sage.

Wright, R. 1994. The moral animal: Why we are the way we are. New York: Vintage.Yankelovich, D. 1982. Changing public attitudes to science and the quality of life: Edited

excerpts from a seminar. Science, Technology, & Human Values 7:23-29.Ziman, J. 1992. Not knowing, needing to know, and wanting to know. In When science meets the

public, edited by B. V. Lewenstein, 13-20. Washington, DC: American Association for theAdvancement of Science.

MICHAEL F. WEIGOLD is an associate professor of advertising at the University ofFlorida. He teaches mass communication theory and advertising research.


SCIENCE COMMUNICATIONBorchelt / COMMUNICATING THE FUTUREThis article reports the principal findings and recommendations of the Research Roadmap Panelfor Public Communication of Science and Technology in the Twenty-first Century. Beginning in1998, the Space Sciences Laboratory at NASA’s George C. Marshall Space Flight Center char-tered a fifteen-member working group to develop a research strategy that would address the bigquestions in science communication academic research and identify the best practices in scienceand technology communication as they are being implemented in research institutions across theUnited States and abroad. The working group met eight times at various U.S. research institu-tions, invited science communicators and others to meet with them, and solicited public andother comment in preparation for this article.

Communicating the FutureReport of the Research Roadmap Panel for Public Communication

of Science and Technology in the Twenty-first Century

RICK E. BORCHELTU.S. Department of Energy

Beginning in 1998, the Space Sciences Laboratory (SSL) of NASA’s GeorgeC. Marshall Space Flight Center (MSFC) established a blue-ribbon panel1 ofscience communicators, communication researchers, Pulitzer Prize–winningjournalists, and scientists to assist its efforts in public communication ofNASA research. The SSL had recently reorganized to support an in-housecommunication function separate from the MSFC public affairs office, oneaimed principally at directly communicating scientific results to lay audi-ences rather than channeling those messages through mass media. Senior

Author’s Note: The work of the Research Roadmap panel reflected in this article was supportedby funding from the George C. Marshall Space Flight Center, Huntsville, Alabama. Much of thetext was drawn from committee writings and as such represents a collective, rather than an indi-vidual, work. Errors and interpretations, however, are solely the author’s. Address correspon-dence to Rick E. Borchelt, U.S. Department of Energy, Director of Communications, Office ofScience (SC-5), 1000 Independence Ave., SW, Washington, DC 20585; phone: 202-586-6702;fax: 202-586-7719; e-mail: [email protected].


194

researchers at the SSL hoped to have the panel review the fundamentalresearch base that underpins public communication of science and technol-ogy, advise them on some areas of communication research that would bene-fit most from funding, and identify best practices in science and technologycommunication that MSFC/SSL might wish to adopt for its own use.

Working under a cooperative agreement with the University of Florida,which had previously conducted a communications audit and other researchon MSFC’s science communication enterprise, the Research Roadmap Panelfor Public Communication of Science and Technology in the Twenty-firstCentury (dubbed the R2 group) met formally eight times from 1998 to 2000,hosted each time by various research institutions across the United States.2 Ateach meeting, science communicators, journalists, scientists, and researcherswere invited to attend to share their experiences in public communicationwith the panel. In addition, an open invitation was issued to the community ofscience communicators (as reflected in membership in the National Associa-tion of Science Writers and other professional organizations, or members ofthe working press covering science and technology in the region) to join thepanel’s deliberations. The meetings were open to the public, and MSFC wasfully committed to open sharing of the panel’s findings and recommendations.

In addition, panel members met informally at major professional meet-ings such as the Council for the Advancement of Science Writing’s NewHorizons Briefings and the annual meetings of the National Association ofScience Writers and the Association for Education in Journalism and MassCommunication.

Panel Charge

The panel was charged with two very different tasks. First, to help guidefuture NASA investments in science communication research, the panel wasasked to review the recent literature on science communication and relatedfields and to frame the big questions that remained to be answered by com-munication scholars working in science and technology communication.Second, the panel was asked to survey science communication activities atscientific research institutions in the United States and abroad for models thatcould be adapted for use at MSFC or other research-performing organizations.

To address the first charge, the panel commissioned a review of the recenttraditional science communication literature. That review, by University ofFlorida associate professor and panel member Michael Weigold (2001 [thisissue]), can be found in this issue of Science Communication. The panel

Borchelt / COMMUNICATING THE FUTURE 195

commissioned a complementary review of the health communication litera-ture from Michael Antecol, then of the Stanford Center for Research in Dis-ease Prevention and directed panel member Robert Logan of the Universityof Missouri to do a cursory review of the literature from agriculture extensionand report back to the group on the potential value of a complementaryreview of that literature. On the basis of these reviews, and in consultationwith other science communication researchers, the panel outlined a series ofquestions that need to be addressed by future research. The panel also spon-sored pilot studies in three of the most promising research areas, and two ofthose studies (Priest 2001 [this issue]; Tremayne and Dunwoody 2001 [thisissue]) appear in this issue of Science Communication.

In addressing the second charge, the panel consulted widely and revieweddozens of existing science communication efforts undertaken by universities,corporations, public relations agencies, museums, professional societies andorganizations, and other science communication practitioners. In addition,the panel organized an international peer-reviewed conference to reviewbest-practice submissions, cosponsored by the U.S. Department of Energyand the National Institute of Standards and Technology (NIST) and sched-uled for 6-8 March 2002 at NIST’s Gaithersburg, Maryland, campus. Entriesselected as best practices were invited to develop poster presentations for theconference that will be archived on the World Wide Web and in print form.This article lays the groundwork for identification of these best practices.

The panel also commissioned a review of federal science communicationactivities from Bruce Lewenstein at Cornell University. That review was notcomplete as this issue went to press.

The Purposes of Communication

While the panel recognizes that many societal needs are fulfilled by com-munication about science and technology, for the purpose of this article, thepanel identifies three primary purposes for the communication of scientificinformation by agencies and institutions. It is communication of these typesthat the panel addresses in its findings and conclusions:

To inform consumers, patients, and citizens about scientific activities,products, or conclusions that may be useful in improving the quality of lifegenerally or in regard to specific problems, issues, or events. This kind ofcommunication would include messages from the National Institutes ofHealth about new medical research, information from the Department of


Agriculture about the safe use of pesticides, reports from the Department ofTransportation about the safety of specific vehicles, or information from theNational Park Service about vacation sites and resources.

To provide information for citizens to enable them to understand, thinkabout, and perhaps participate in the formulation of public policy on specificissues. This kind of communication might include information from theDepartment of Agriculture and the Food and Drug Administration aboutgenetically modified foods, information from the Department of Energyabout current energy resources and future needs, or information from NASAabout the status and uses of the space station. Some of these communicationsmay be persuasive in character, while others may involve only a presentationof research results. Some messages in this category may involve a compila-tion of differing views, options, and arguments.

To provide descriptions and explanations of scientific work to enhance thelevel of scientific or biomedical literacy in the recipient. This kind of commu-nication is represented by the programs of museums, agency visitor centers,and Web sites to provide new information about previous and current scien-tific work. The numerous Web sites operated by NASA, the National Insti-tutes of Health, the Smithsonian, and numerous agencies, universities, and pro-fessional societies provide additional examples of this kind of communication.

Principal Findings and Conclusions—Research Roadmap

In his literature review in this issue, Weigold (2001) provides a moreextensive roadmap and rationale for a research agenda in science communi-cation than is possible here. The panel endorses these recommendations.

In particular, the panel notes that as the disciplines of science communica-tion per se and health communication have matured, many academicians havechosen one field as a specialty to the exclusion of the other. This has led to thedevelopment of two very distinct fields of endeavor that have lost much oftheir potential for interdisciplinary collaboration and mutual cross-fertilization.

Another related discipline with promise for informing science communi-cation is that of agricultural communication. An early communicationmodel, diffusion of innovations, was developed to explain the processwhereby agricultural innovations diffuse through opinion leaders and earlyadopters to the broader population. Today, agricultural communication


scholars are dealing with issues of direct relevance to science communica-tion, including the public’s acceptance of genetically modified foods.

To make the most of limited resources to support academic research in thebroader field of public communication about science, health, and technology,the panel urges the research community and the funding community for theseendeavors to actively seek opportunities for greater collaboration and syn-chronization of research.

The panel identified three research areas as being especially deserving ofattention from science communication scholars and their allied colleagues:

Exploring the Relationship between Quality orQuantity of Science Communication, AdultScientific Literacy, and Citizen Science Advocacy

The panel noted that much of current communication practice assumesthat good science communication yields benefits in terms of broader citizensupport for the scientific enterprise. This assumption is not supported in theliterature, and in the panel’s view, it is unlikely to accurately represent what isclearly a very complex communication system.

This issue is not unique to science communication. To take one example,political communication scholars dating back to the nineteenth century havedecried the ignorance of voters about their political environments, politi-cians, issues, and events. This ignorance flourishes despite the fact that politi-cal topics receive substantially heavier coverage in the news media than doesscience and despite the fact that the reporters who cover politics generally aresophisticated and knowledgeable about their topic. These reporters are wellrewarded and represent some of the best talent in the news industry. And,those who produce political messages are among the most skilled: politiciansand their press officers often are unrivaled experts at message packaging andpresentation (in stark contrast to common portrayals of scientists). In otherwords, in politics, the public receives a large amount of news by expertreporters interviewing the masters of sound bites. Yet, people frequently can-not name both of their senators, have no idea who the nine justices on theSupreme Court are (or even that there are nine justices), and in general claimto lack respect for elected officials and the people who cover them.

The panel notes these things to make a simple point: political ignoranceflourishes in spite of heavy coverage, knowledgeable reporting, andmedia-savvy participants. In addition, the public’s evaluation of both report-ers and politicians is not especially positive. There may be some lessons and


important cautions from the broader field of political communication as weaddress prescriptions for enhancing science communication.

Understanding the Interests and Behaviors of PublicsWho Consume or Use Science and Technical Information

Similarly, much of current public communication practice is based onidentifying what the public ought to know and providing that, rather thanidentifying what the public wants to know and finding ways to make thisknowledge available and accessible (Ziman 1992). While there is a rich liter-ature on uses of political information (see, e.g., McCombs and Shaw 1993),as well as on health information and health-behavior modification, the corre-sponding literature on use of scientific information per se is relatively slight.

Understanding the Ecosystem ofCommunication between the Research Scientist,the News Reporter or Other Communicator,and the Institutional Public Information Officer

Little is known about the role that the science public information officer(PIO) plays in brokering communication between scientists and representa-tives of the media and other external audiences, in part because the definedcareer of this individual is a very recent phenomenon. However, the panelbelieves that the role of the PIO is expanding rapidly, both in scope and in thenumber of institutions and organizations that employ him or her, and that therelationships that a PIO manages with external audiences increasingly influ-ence the process and products of science communication.

One promising model for studying this ecosystem may be game theory.Originally developed in the 1950s, game theory models how actors chooseamong behavioral options as a function of the rewards and costs for suchchoices. It goes beyond simple stimulus-response formulations, however,because it can model the actions of two or more decisionmakers simulta-neously and can develop such models based on whether the decisionmakershave full or incomplete information about their choices. Game theory hasbeen applied productively to politics, international relations, and social rela-tionships. It has recently become more important in public relations. Gametheory may provide a useful way to study the changing dynamics in sciencecommunication because it provides a way to predict how decisionmakers willrespond to changing reward-cost structures in their environments.


Principal Findings and Conclusions—Best Practices

The panel was struck overall by the general lack of intellectual rigorapplied to science and technology communication activities, especially ascontrasted with the very rigorous scientific environment in which this com-munication arises. Communication often remains an afterthought, a by-prod-uct of scientific endeavor somehow removed from the scientific process itselfand often funded by a different mechanism than the scientists who performthe research. The panel firmly believes that public communication ofresearch results is, and should be, integrated into the scientific process itself.It is not an optional activity at the conclusion of a research program. It shouldbe amenable to the same experimental paradigms as laboratory science’s.

The panel also was very concerned about the dearth of formative orevaluative research that underpins the vast majority of science and technol-ogy communication in the United States (and as far as the panel was able todetermine, the rest of the world). For a data-driven enterprise, sciencedemands very few data from communicators of science, either to craft andframe appropriate messages and message content or to evaluate the impact ofmessages on scientific knowledge or behavior. The best evaluation seems tooccur in the context of health-behavior campaigns, where the end product is adefinable set of behavioral outcomes. As a rule, this kind of evaluative frame-work is lacking entirely in communication programs about basic researchand technology. The panel urges science communicators to undertake rigor-ous formative and evaluative research as part of any communication process.

One last general observation concerns the role of mass media in nurturingpublic understanding of science and technology. As a rule, the panelobserved, mass media are a very poor tool for remedial science education.Basic understanding of science and technology is only minimally affected inadult life by consumption of media stories about scientific issues. It appearsthat the role of K-12 education is far more important than subsequent expo-sure to science communication (Friedman, Dunwoody, and Rogers 1999).

The remainder of the panel’s principal findings with respect to best prac-tices in public communication of science and technology fall under six gen-eral observations.

• Finding 1: There is no such thing as a general audience for science and technol-ogy communication; rather, there are many people with many different uses forscience and technology information and many levels of understanding withwhich to deal.


The public is not a uniform whole but is segmented by differing interests,differing abilities, differing resources, and differing needs. This is not a newidea. In the tenth Federalist Paper, James Madison wrote that one of thestrengths and protections of a democratic society is the plurality of interestsfound in each citizen.

The research programs of scholars such as Miller (1986) and Prewitt(1982) have helped immensely in understanding this plurality of interestsamong science news consumers. Among their somewhat bleak findings arethe following: almost half of American adults report that they do not followany public policy area closely (Miller 1986). Prewitt (1982) and Miller(1986) both advocate segmenting science news audiences based on theirinterests in science. Heavy consumers of science news are a minority ofadults, but they remain important because of their prominence and impor-tance in society.

The most recent assessment of U.S. attitudes about science and technol-ogy, the National Science Board’s (NSB’s) (2000b) Science and EngineeringIndicators 2000 report, found that less than 10 percent of the U.S. public canbe considered “science attentive” for most issues covered by the NSB survey.Science attentives are those individuals who express a high level of interest ina particular issue, feel well-informed about that issue, and read regularlyabout that issue. Medical research has the largest audience, about 16 percent.Similarly, few Americans are likely to be attentive to science and technologypolicy issues—about 12 percent. The “interested” public—those who claim ahigh level of interest but do not feel well informed about a particulartopic—comprises about 44 percent of the population. Miller (1986) charac-terized the demographics of these populations, noting that the science-atten-tive public is more likely than the population at large to be younger, male,better educated, and to have taken a college-level science course. Sci-ence-interested publics are older, less educated, and less likely to have had acollege-level science class.

There is a wide disparity in the kinds of science and technology informa-tion generally known by the U.S. population. More than 70 percent of Ameri-cans know, for example, that oxygen comes from plants, that the continentsare moving and have done so for millions of years, that light travels fasterthan sound, and that the Earth goes around the Sun. However, one-half orfewer of Americans know that the earliest humans did not live at the sametime as dinosaurs, that it takes the Earth one year to go around the Sun, thatelectrons are smaller than atoms, or that antibiotics do not kill viruses (NSB2000b).


Despite long-standing awareness of the diversity of the consumingpublics for science and technology information, the panel noted that mostscience communication still fell into one of only two categories: peer com-munication aimed at fellow scientists and technologists and public communi-cation aimed at everyone else. The literature the panel reviewed and the bestpractices it observed in use make very clear that there is no such thing as aone-size-fits-all public communication message for a mythical lay public.Single messages designed to reach all public audiences typically end upreaching none of them very well, especially in an information environmentwith a myriad of media channels (which is growing daily) from which anaudience may choose what suits it.

This finding flies in the face, also, of traditional mediated communicationprograms, which see their principal or only focus as delivering news items upto the news media. While mediated communication has an important role toplay in increasing public understanding of science and technology (and thisrole will continue in the foreseeable future, the panel believes), public dis-course is no longer driven by a few major media players. An individual articleor story placed in an individual news medium is more likely to be lost in thevery crowded intellectual marketplace than it is to have a profound impact onpublic understanding of science.

All communication should have an intended audience, and most messagesare designed to be received and used by selected individuals and groups. Theprior selection of an audience is important because audiences differ in theirinterests and in their ability to use various kinds of information. The prepara-tion of a one-size-fits-all message for all possible audiences and outlets isalmost always ineffective and is a practice to be discouraged.

The effectiveness of communication—the accurate receipt and use ofinformation—can be improved substantially by carefully defining intendedaudiences and by tailoring the level of information provided to each audi-ence. While many federal agencies and grant-receiving institutions feel that itis necessary in a democratic society to provide all public information in astyle and format accessible to adults with an upper-elementary reading level,it is important to recognize that citizens differ in interests, in their level ofeducation and scientific literacy, and in the amount of time and effort they arewilling to devote to any given subject or issue. Some individuals will prefer(and more effectively utilize) written material while other citizens may preferand need pictures, audio, and graphic presentations.

The panel also notes in this context that extensive reliance on general-public messages seriously undermines efforts to address hard-to-reach audi-ences such as racial and ethnic minorities and those without Internet access.


The panel recommends that federal agencies and their grantees and con-tractors recognize the multifaceted nature of the public (individuals, groups,and institutions) and design communication programs appropriate to theneeds of each group. This approach should mean not that some groups areinherently better served than others but that the type and level of communica-tion is designed to address and serve the needs of each group within thepublic.

• Finding 2: The scientific community and managers of the science enterpriseroutinely fail to distinguish between understanding of science and appreciationfor science- or research-performing institutions.

The panel believes that both of these goals are appropriate and laudableunder the right circumstances. But far too often, the panel observed, commu-nication programs that are intended to enhance the reputation and cachet ofindividual agencies, institutions, or organizations are touted as programs thatincrease public understanding of science. The goals of these two programsare not necessarily complementary and in fact often work at cross-purposes.

In particular, the panel notes that collaboration is essential to the processof science—professional collegiality undergirds the infrastructure of scien-tific research. But, institutional reputations are made and preserved by claim-ing credit for scientific advancements or technological achievements, andsharing of credit dilutes institutional advancement goals.

Moreover, the effective communication of the process of science (whichthe panel believes is equal in importance to communication about products ofscience if the goal is public understanding) requires an acknowledgementthat scientific experiments do not always work and that this kind of failure isas instructive and valuable as experiments that yielded the expected results.But scientists working with public dollars often are reluctant to discloseresearch failures, leading to unrealistic public expectations about scientificprogress. Such failure is generally seen as anathema to institutionaladvancement.

The lack of distinction between these sometimes-competing goals alsoleads to poor metrics of communication efficacy. While evaluation is gener-ally poor across the board in science communication (as noted above), wheremetrics do exist, they are more likely to be measures of approval or supportthan measures of knowledge or behavior.

The panel recommends that communication campaigns in science andtechnology explicitly address at the outset whether the goal is public under-standing or public appreciation, and design metrics appropriate for measur-ing the desired outcome.


• Finding 3: Science and technology communication should not be driven by theresearch enterprise’s desires about what the public should know. Communica-tion should be driven by a desire to meet audience needs and interests.

Scientists have an obligation to understand publics and their communica-tion needs if communication is to be effective. Once again, the multiplicity ofavailable media channels makes it unrealistic to expect an audience to attendto messages or communication in which it has no interest. There are no cap-tive audiences for science and technology information.

• Finding 4: The active involvement of scientists and engineers is critical to thesuccess of science communication.

In 1996, Neal Lane (cited in Cialdini 1997), then director of the NationalScience Foundation, challenged scientists by suggesting, “If you don’t take itas one of your professional responsibilities to inform your fellow citizensabout the importance of the science and technology enterprise, then the pub-lic support—critical to sustaining it—isn’t going to be there” (p. 676).Although a direct causal relationship between communication and agencyfunding is doubtful, many voices overwhelmingly suggest that principles ofpublic accountability will require researchers will be expected to describewhat society has received for its investment.

Most academic research suggests that in general, scientists are interestedin educating the public through the mass media, they understand that theyhave such an obligation (DiBella, Ferri, and Padderud 1991), and they arewell aware of the possible advantages of doing so (Dunwoody and Ryan1983; Nelkin 1995). While scientists are wary that communicating to the laypublic extends their accountability beyond the scientific community, cooper-ation among scientists and journalists appears to be growing. This coopera-tion is occurring in spite of the well-documented differences and problemsbetween the two cultures.

While the previous discussion suggests that scientists, by and large, dounderstand that they have an obligation to educate the public about science,the panel believes that this attitude needs to become more pervasive. Scien-tists need to understand that to fulfill this public service obligation, they mustinteract with the media and other publics external to the peer community.While it is clear that some scientists naturally will be better communicatorsthan others, all scientists have a stake in and obligation to the outcomes ofpublic communication. This obligation may be as minimal as responding toor providing information for a reporter when he or she is contacted, or it may


involve more central involvement in an organization’s public communicationprograms.

In particular, the panel finds that scientists need a working knowledge, or“culture appreciation,” of the media and how they operate. As Nelkin (1995)suggested, “scientists and journalists must accept and come to terms with[emphasis added] an uneasy and often adversarial relationship” (p. 171). Sci-entists should be taught, through communication training sessions, to recog-nize that science communication is a field that is backed by rigorous researchand strong professional standards. Scientists should also learn that science jour-nalists share common goals of accurate and fair information dissemination.

This public service mind-set needs to be extended to include other efforts.The panel endorses the NSB recommendation in its report on communicatingscience (NSB 2000a) that scientists and engineers need

to be more articulate and clear about their work and the good it is doing for soci-ety, be more accessible and more accountable, and lead or participate in publicinformation efforts in a wide variety of public forums—from schools to themedia. (P. 17)

The panel also echoes the NSB’s admonition to scientists and engineers to“communicate the joy and fascination of science as well as its utility” (NSB1998, 15).

Most academic research and expert advice gathered from the panel’smeetings suggest that the active involvement of scientists and engineers at theorganizational level also is critical to the success of any science communica-tion endeavor. The panel believes that the most effective science organiza-tions are those that integrate scientists in joint and equal decision makingregarding science communication issues, including the content and timeframes for release of information. In this scenario, the importance of sciencecommunication permeates the entire culture of the organization; organiza-tional leaders place utmost value in this activity.

It must be recognized that organizations for which this cultural shift hasbeen most successful have put institutional reward systems in place. Thepanel recommends that scientists be rewarded for aiding in the public com-munication efforts of their organization. These rewards can range fromnonmonetary recognition in the form of awards or in-house newsletter arti-cles to more traditional monetary rewards. In practice, it would be wise fororganizational leaders to solicit input from scientists about appropriate andmeaningful rewards for these activities.


• Finding 5: Science communicators who can foster mutual respect between sci-entists and external publics are essential to effective public communication ofscience.

At a recent conference in London, science historian Bruce Lewenstein(2000) of Cornell University traced the origins of communication with thepublic about science by science institutions back to the genesis of public sci-ence museums in the United States at the end of the nineteenth century. Thesenascent efforts were followed by the rise of scientific societies like the Amer-ican Medical Association and the American Chemical Society, which begancoordinated campaigns to convince the public of the benefits of science asearly as 1910 to 1915. With the formation of the National Association of Sci-ence Writers in 1934, the communication of science to the public began anevolution from a conscious effort to show the value of science to a moreobjective, less value-laden reporting of scientific advances that continues toshape American journalistic coverage of science today. By the middle of thetwentieth century, noted Lewenstein (2000), public communication aboutscience had emerged as a career in and of itself.

At the same time, though, Lewenstein (2000) pointed out that the nature ofscience communication activities for the public were determined more by theparticular goals and concerns of dedicated individuals—who moved freelyamong private, commercial, educational, and government positions—thanby particular institutions.

Nelkin (1995) noted that despite the growing number of science commu-nicators working for institutions, as recently as twenty-five years ago, institu-tional science communication was a field many people “fell into” rather thanconsciously chose as a career. Training was typically on the job, and therewere few opportunities for meaningful professional development. In 2000,new entrants to the field of institutional science communication were muchmore likely to have been trained for the profession by earning a degree or cer-tification from one of several dozen U.S. colleges or universities offering spe-cialty studies in this area (Dunwoody et al. 1998).

Most scientific institutions are decentralized, with a relatively flat organi-zational structure. Owing to the specialized nature of science, employees ofresearch-performing institutions usually are well educated, and day-to-daydecision making occurs at relatively low levels of the organization. In thisenvironment, the science communicator with typically a bachelor’s or mas-ter’s degree in journalism, English, or science communication is often theodd person out. Especially at the senior levels of most scientific organiza-tions, the director of the institution’s public affairs or communication officeoften is one of the few non–Ph.D. trained executives sitting at the table.


To speak with authority under such circumstances, the director of publicaffairs ideally should report to the head of the agency, department, ormuseum or to the president of the professional society or university. Short ofa direct reporting relationship, the director of public affairs needs unfetteredaccess to the head of the institution on very short notice. Communicationdecisions typically must be made within very short time frames. Manyreporters have daily deadlines, and responses must be developed and commu-nicated quickly. A public affairs director must have easy access—and prefer-ably a direct reporting relationship—to the head of the organization toaccomplish timely and informed decision making in responding to mediainquiries.

Providing a seat at the table for public affairs also helps ensure that theorganization will be better able to consider the public consequences of itsactions. Should a minor chemical spill be reported to the surrounding com-munity now or only if a reporter asks about it? Should the organization’s Webpages project a united front to Web surfers, or should each of the organiza-tion’s divisions be free to develop its own format? Should an inquiry from acongressional committee about research facilities be handled by building ser-vices experts or by the head of the organization? A forceful public affairsdirector often will answer such questions differently than will a Ph.D. scien-tist. Without a seat at the table, the public perspective is often lost, and theorganization makes less-informed decisions.

The panel lauds the trend toward professionalization of science communi-cation. It recommends that science communication professionals in research-performing institutions participate meaningfully as part of the organization’ssenior management. How that relationship is developed and implemented ishighly dependent on the nature and structure of the institution, and the panelhas reviewed exemplary practices that include direct reports, institutionalleadership, dual reporting roles, and the leaders of an institution’s researchfunction and its overall leadership.

• Finding 6: The proliferation of new media and the fragmentation of existingmedia will have profound impacts on how and with whom one communicatesabout science and technology.

Even given the recent downturn in the fortunes of Internet-based dot-comcompanies, the trajectory of growth of Internet use in comparison to othermass media is impressive. Stempel, Hargrove, and Bernt (2000) found thatsubscribing to Internet and online services increased dramatically and listen-ing to radio news and talk shows increased significantly among Americans innational surveys conducted in 1995 and 1999. They found, during the same


period, significant declines in watching local and network television newsand in reading daily newspapers, grocery store tabloids, and newsmagazines.

In February of 2001, the Pew Internet & American Life Project (Rainieand Packel 2001) estimated that more than 168 million Americans (56 per-cent) had World Wide Web access from either home or work. On a typical dayat the end of 2000, 58 million Americans were logging on—an increase of9 million from the daily figures just six months previous. The online popula-tion is skewed toward the young: fully 75 percent of those between ages eigh-teen and twenty-nine have Internet access, compared with only 15 percent ofthose ages sixty-five and older. Moreover, there are significant differences inonline access by income, with 82 percent of those living in households withannual incomes of more than $75,000 having access compared with only 38percent of those in households earning less than $30,000.

The Pew report also noted that the average American user at the end of2000 spent slightly more than four hours a week online; almost half of theironline time was devoted to e-mail, with the remainder pretty much evenlydivided between browsing for fun or hobby information and getting news—often health news. This is less than the several hours per day that many adultswatch commercial television, but recent Nielsen ratings (data taken from theonline Nielsen NetRatings page for the week of 22 July 2001: http://www.nielsen-netratings.com) suggest that Internet use is increasing at the expenseof attentiveness to other mass media. The Nielsen data put the estimate of the“current Internet universe” of users at 167 million in July 2001.

Despite the growth and robustness of this medium, the panel does notbelieve, however, that Internet-based science communication will be the onlymedium of public communication in the future. Books, magazines, journals,newspapers, and broadcasts continue to be important mechanisms for dis-seminating scientific information to public audiences, and some of thesemedia—notably broadcast cable—are experiencing rapid growth as well.What the Internet offers is unparalleled opportunity to directly reach audi-ences of import, especially the science-attentive and science-interested audi-ences described above. Moreover, the Internet allows direct interaction withscientists and the scientific process in a way difficult to replicate with staticmedia—even though very few communicators take full advantage of thiscapability.

The panel recommends that scientific organizations manage where practi-cable a diverse science communication portfolio. Furthermore, each organi-zation should develop Internet-based science communication programs thatmake full use of the World Wide Web’s ability to reach individual usersdirectly, rather than through mass media, and that take maximum advantageof the Web’s interactive qualities.


Conclusion

For half a century after World War II, the U.S. scientific enterprise thrivedin the context of military preparedness and economic competitiveness withrespect to Russian-bloc countries. The end of the Cold War at the conclusionof the past century, however, forced scientists to begin to think more broadlyabout other societal justifications for research and development and to beginto examine the value of science as a public enterprise. Coupled with stagnantor dwindling fiscal resources since the early 1990s (in constant dollars), thisnew environment has given an unprecedented prominence to the practice andpractitioners of science communication. Unfortunately, other pressures atwork in the scientific enterprise—commercialization, economic competi-tion, and dwindling resources—are dictating the nature and scope of sciencecommunication in ways that the panel believes may not be fully consistentwith better public understanding of science and technology. Rather, manyresearch-performing institutions are adopting marketing, branding, andadvertising approaches that may work well in a commercial enterprise butthat seldom make an effective substitute for good science communication.

The new century will provide many new opportunities to increase publicunderstanding of science and technology. However, these new opportunitiesmust be based on sound scholarship and evaluation—commodities that thepanel finds in very scarce supply among science communication programs inthe United States today.

Leaving aside all potential benefits of science communication to increasepublic advocacy and support for research, the panel believes that better publicunderstanding of science and technology—aided by appropriately designedscience communication programs—is a worthy goal in and of itself. The pub-lic dialogue that results from effective science communication can be a hall-mark of citizen involvement in science for the twenty-first century and thebest possible outcome of communication strategies aimed at better publicunderstanding of science, technology, and health.

Notes

1. Panel members included Rick E. Borchelt (chair), U.S. Department of Energy; DebbieTreise (study director), Department of Advertising, University of Florida; Deborah Blum,School of Journalism and Mass Communication, University of Wisconsin–Madison; LynneFriedmann, Friedmann Communications; Martin Glicksman, Department of Materials Sciencesand Engineering, Rensselaer Polytechnic Institute; John M. Horack (ex officio), Space SciencesLaboratory (SSL), George C. Marshall Space Flight Center (MSFC); Robert Logan, School ofJournalism, University of Missouri; Paul Lowenberg, Lowenberg Communications; Charles


McGruder III, Department of Physics and Astronomy, Western Kentucky University; Jon D.Miller, Northwestern University Medical School; Gail Porter, National Institute of Standardsand Technology; Carol L. Rogers, College of Journalism, University of Maryland; BarbaraValentino, Evolving Communications; Michael Weigold, Department of Advertising, Univer-sity of Florida; Gregory Wilson (ex officio), SSL, MSFC; and Kris Wilson, Department of Jour-nalism, University of Texas.

2. Meetings were held at The Salk Institute, La Jolla, CA; the Marine Biological Laboratory,Woods Hole, MA; Duke University, Durham, NC; the American Association for the Advance-ment of Science, Washington, DC; NASA MSFC, Huntsville, AL; Northwestern UniversityMedical School, Chicago; University of California, Santa Cruz; and the University of Florida(meeting held in Jacksonville, FL).

References

Cialdini, R. 1997. Professionally responsible communication with the public: Giving psychol-ogy a way. Personality and Social Psychology Bulletin 23:675-83.

DiBella, S. M., A. J. Ferri, and A. B. Padderud. 1991. Scientists’ reasons for consenting to massmedia interviews: A national survey. Journalism and Mass Communication Quarterly68:740-49.

Dunwoody, S., D. Harp, R. Barth, and E. Crane. 1998. Directory of science communicationcourses & programs in the United States. Madison: University of Wisconsin Board ofRegents. Available: http://www.journalism.wisc.edu/dsc/index.html.

Dunwoody, S., and M. Ryan. 1983. Public information persons as mediators between scientistsand journalists. Journalism Quarterly 60:647-56.

Friedman, S., S. Dunwoody, and C. Rogers, eds. 1999. Communicating uncertainty: Media cov-erage of new and controversial science. Mahwah, NJ: Lawrence Erlbaum.

Lewenstein, B. 2000. An American historical perspective on public communication of science.Paper presented at the Conference on Science Communication, Education, and the Historyof Science, London, 12-13 July.

McCombs, M. E., and D. S. Shaw. 1993. The evolution of agenda-setting research: Twenty-fiveyears in the marketplace of ideas. Journal of Communication 43:58-67.

Miller, J. 1986. Reaching the attentive and interested publics for science. In Scientists and jour-nalists: Reporting science as news, edited by S. M. Friedman, S. Dunwoody, and C. L. Rog-ers, 55-69. New York: Free Press.

National Science Board (NSB). 1998. The National Science Board strategic plan. NSB 98-215.Washington, DC: Government Printing Office.

. 2000a. Communicating science and technology in the public interest. NSB 00-99.Washington, DC: Government Printing Office.

. 2000b. Science and engineering indicators 2000. NSB 001. Washington, DC: Govern-ment Printing Office.

Nelkin, D. 1995. Selling science: How the press covers science and technology. Rev. ed. NewYork: Freeman.

Prewitt, K. 1982. The public and science policy. Science, Technology, & Human Values 36:5-14.Priest, S. H. 2001. Misplaced faith: Communication variables as predictors of encouragement

for biotechnology development. Science Communication 23:97-110.Rainie, L., and D. Packel. 2001. More online, doing more. Washington, DC: Pew Internet &

American Life Project. Available: http://www.pewinternet.org.


Stempel, G. H., III, T. Hargrove, and J. P. Bernt. 2000. Relation of growth of use of the Internet tochanges in media use from 1995 to 1999. Journalism & Mass Communication Quarterly 77(1): 71-79.

Tremayne, M., and S. Dunwoody. 2001. Interactivity, information processing, and learning onthe World Wide Web. Science Communication 23:111-63.

Weigold, M. F. 2001. Communicating science: A review of the literature. Science Communica-tion 23:164-94.

Ziman, J. 1992. Not knowing, needing to know, and wanting to know. In When science meets thepublic, edited by B. V. Lewenstein, 13-20. Washington, DC: American Association for theAdvancement of Science.

RICK E. BORCHELT is the director of communications for the U.S. Department ofEnergy’s Office of Science. A naturalist by training, he was press secretary for the WhiteHouse Office of Science and Technology Policy during the Clinton administration andpress secretary for the Science, Space and Technology Committee of the U.S. House ofRepresentatives under the chairmanship of Representative George E. Brown, Jr. Hiscareer also has spanned science communication posts at the National Academy of Sci-ences, the University of Maryland, NASA, Lockheed Martin, and Vanderbilt University.


SCIENCE COMMUNICATIONNEWS AND NOTICES

NEWS AND NOTICES

AAAS To Meet February 2002

Wide-ranging symposia, plenary and topical lectures, seminars, careerworkshops, and an exhibit are all part of the activities planned for the Ameri-can Association for the Advancement of Science (AAAS) Annual Meetingand Science Innovation Exposition scheduled for 14-19 February 2002 inBoston. Several thousand scientists, engineers, and others interested in sci-ence are expected to attend the meeting, as are hundreds of journalists fromaround the world.

Among the themes around which symposia will be organized are thefollowing:

• communicating across boundaries;• cultural and social diversity;• global change;• environmental and biological diversity;• governing science and science in government;• science and society;• science, engineering, and public policy;• science and the public trust;• science and sustainability; and• teaching, learning, and careers.

Specific symposia topics include best practices from research scientistswho communicate with the public; accommodating interdisciplinarity in theacademic research environment; international trends in the transfer of aca-demic research; integrating the science, economics, and policy of climatechange; biodiversity science and global research; science, ethics, and com-munication: shaping public policy of low dose radiation; managing the aca-demic research enterprise; stewardship of digital scientific information;social and ethical implications of behavioral genetics research; agendas andmodes of cooperation in international science; bridging the gap from scien-tific discovery to economic growth; biotechnology policy in Europe andNorth America; public participation in decision making for science, technol-ogy, and society; communicating about global sustainability; citizens’ con-sensus conferences and genetically modified foods; shattering the glass ceil-ing for women in science; the role of undergraduate participation in research;


212

and gender in science, mathematics, and engineering faculty hiring andretention.

For additional information about the meeting, including details about reg-istration and hotel information, contact the AAAS Meetings Department,1200 New York Avenue, NW, Washington, DC 20005; phone: 202-326-6450;fax: 202-289-4021; e-mail: [email protected]; or visit the AAAS Website: http://www.aaas.org/meetings.

Cheiron Call for Papers

Scholars working in areas related to the history of the behavioral andsocial sciences or with related historiographical and methodological issuesare invited to submit papers, poster abstracts, symposia abstracts, or propos-als for workshops for the thirty-fourth annual meeting of Cheiron, the Inter-national Society for the History of Behavioral and Social Sciences. All sub-missions must be received by 14 January 2002. The meeting will be held26-30 June 2002 at the University of Oregon, Eugene.

A limited number of student travel awards will be available for studentswho are presenting papers.

Cheiron, named after the centaur of Greek mythology, was formed in 1968to promote international cooperation and multidisciplinary studies in the his-tory of the social and behavioral sciences. The organization publishes a bian-nual newsletter, holds an annual meeting in June of each year, and is affiliatedwith the Journal of the History of the Behavioral Sciences.

For more information about program submissions, contact the CheironProgram Chair, Hans Pols, Institute for Health, Health Care Policy, andAging Research, Rutgers University, 30 College Ave., New Brunswick, NJ08901; phone: 201-330-1449; fax: 732-932-6872; e-mail: [email protected]; or visit the meeting Web site: http://www.psych.yorku.ca/orgs/cheiron/.

European Social Science HistoryConference Set for The Hague

The Fourth European Social Science History Conference will be held atThe Hague, the Netherlands, 27 February–2 March 2002. These biennialmeetings are designed to bring together scholars interested in explaining his-torical phenomena using the methods of the social sciences. Papers and ses-sions are presented on a wide variety of topics and covering diverse historicalperiods.

NEWS AND NOTICES 213

The European Social Science History Conference is organized by theInternational Institute of Social History, an institute of the Royal NetherlandsAcademy of Arts & Sciences. For further information about the conference,contact the Conference Secretariat ESSHC, c/o International Institute ofSocial History, Cruquiusweg 31, 1019 AT Amsterdam, the Netherlands;phone: 31 20 66 858 66; fax: 31 20 66 541 81; e-mail: [email protected]; or visitthe conference Web site: http://www.iisg.nl/esshc.

Technotopias Conference Set for 10-12 July 2002 in Glasgow

Technotopias: Texts, Identities, and Technological Cultures is the themeof an interdisciplinary conference to be held 10-12 July 2002 in Glasgow,United Kingdom. Organized by the Department of English Studies of theUniversity of Strathclyde, the conference aims to “reflect upon the place ofthe arts within modern academia; . . . investigate the complex historical and con-temporary interplay between the humanities and technology; and . . . addressthe impact of these relationships upon the formation of physical and culturalidentities.”

Papers for the conference are invited from all fields of literary and culturalcriticism, as well as from scientific and technological disciplines. Abstractsof papers to be presented are due by 31 December 2001 to [email protected]. Topics might include the following:

• literatures of technology: historical contexts;• frontiers of the imagination: science and fiction;• (post)modern texts/(post)industrial spaces;• technologos: technology and the word;• the science of Angellica: gender and technology;• culture, technology, and the body;• technologies and the self;• new media, old academe; and• paradigms of utility in academia.

For further information about the conference, contact the TechnotopiasOrganizing Committee, Department of English Studies, University of Strath-clyde, Livingstone Tower, 26 Richmond Street, Glasgow G1 1XH UK.


History of Mathematical SciencesConference Set for 20-23 December 2001

A series of invited plenary talks and contributed paper presentations willbe featured at the First International Conference of the New Millennium onthe History of the Mathematical Sciences from 20-23 December 2001 inDelhi, India. Organized by the Indian Society for History of Mathematics,Ramjas College, University of Delhi, and other national institutions, the con-ference will be held at Ramjas College, University of Delhi.

The conference will cover all aspects of the history of mathematical sci-ences, including mathematics, statistics, operations research, computer sci-ence, and those topics’ applications to societal needs. Among the specificareas the conference will address are general histories, source books, andbiographies of mathematicians; mathematics and indigenous cultures of theworld; the origin of mathematics; the history of mathematics as a subject ineducational curricula; and future prospectives.

For additional information about the conference, contact Professor Y. P.Sabharwal, Organizing Secretary, ICHM2001, Department of Mathematics &Statistics, Ramjas College, University of Delhi, Delhi 10007, India; e-mail:[email protected] or [email protected].

Literature and Science Conference Set for Denmark

Scholars from human, social, medical, technical, and natural sciences, aswell as artists, are invited to participate in the Second European Meeting ofthe International Society for Literature and Science (SLS) on 8-12 May 2002at the University of Aarhus, Denmark. The conference is expected to be ofparticular interest to scholars interested in interdisciplinary and transdisci-plinary approaches to and linkages between the study of culture, literature,visual arts, and technoscience and between science and the arts.

For additional information, contact SLS c/o Randi Markussen, AssociateProfessor, Department of Information and Media Studies, University ofAarhus, Niels Juels Gade 84, 8200 Aarhus N, Denmark; phone: 45 89 42 1966; fax: 45 89 42 19 52; e-mail: [email protected]; or visit the conference Website: http://imv.au.dk/SLS-Europe.

NEWS AND NOTICES 215

Video Technology Recommended forInternational Comparative Research in Education

“Video technology has evolved into a powerful methodological tool forinternational comparative research in education,” concludes the Board onInternational Comparative Studies in Education (BICSE) of the U.S.National Research Council in its recent report, The Power of Video Technol-ogy in International Comparative Research in Education (2001, NationalAcademy Press). In particular, videotapes of classroom practices can beespecially useful to highlight teaching practices that are difficult to describeand interpret because of differing languages and cultural contexts, accordingto the board.

Recognizing that many questions remain about both the methodology andpractical applications of using video technology, the BICSE report includesfour recommendations for researchers, funding agencies, and policymakers.It called on the international comparative education research community todo the following:

• pursue projects that appropriately use video technology as a research tool;• support not only large-scale studies that make use of video technology, such as

Third International Mathematics and Science Study (TIMSS), but also otherkinds of video-based research;

• undertake initiatives, such as the support of a working group to help clarify anddevelop solutions to the privacy and confidentiality issues in using video tech-nology in such research; and

• undertake initiatives, such as the support of a working group, to explore the cre-ation of a videotape archive or archives for international comparative researchin education.

BICSE provides guidance on the conduct and use of TIMSS, as well asother large-scale education studies, and provides comparative analysis ofeducation systems. The report on the use of video technology was developedthrough a series of meetings and subsequent activities, inspired in part by thepublic release of the TIMSS Videotape Classroom Study.

The Power of Video Technology in International Comparative Research inEducation, by the Board on International Comparative Studies in Education,Monica Ulewicz and Alexandra Beatty, editors (Washington, DC: NationalAcademy Press, 2001, 44 pages, ISBN 0-309-07567-X). For further informa-tion, visit the NAP Web site: http://www.nap.edu.


science communication - sage publications · science communication an interdisciplinary social...

Documents